ci(build): use docker.io package instead of docker-ce in build jobs

The Debian Bullseye runner image doesn't have the Docker CE
repository configured. docker.io is available from default repos.

.env.example

@@ -0,0 +1,75 @@
+# SPARC Configuration
+
+# ---- Application Environment ----
+# Set to "production" or "staging" in deployed environments.
+# The API will refuse to start with the default JWT secret unless APP_ENV=development.
+APP_ENV=development
+
+# ---- API Keys ----
+
+# SerpAPI key for patent search
+API_KEY=your_serpapi_key_here
+
+# OpenRouter API key for LLM analysis
+OPENROUTER_API_KEY=your_openrouter_key_here
+
+# ---- Database ----
+
+# PostgreSQL credentials (used by docker-compose)
+POSTGRES_USER=postgres
+POSTGRES_PASSWORD=change-me-to-a-secure-password
+POSTGRES_DB=sparc
+
+# Full database URL (must match the credentials above)
+DATABASE_URL=postgresql://postgres:change-me-to-a-secure-password@localhost:5432/sparc
+
+# ---- Authentication ----
+
+# JWT Secret for signing tokens
+# IMPORTANT: Change this to a secure random string in production
+JWT_SECRET=your-secure-jwt-secret-change-in-production
+
+# ---- CORS ----
+
+# Comma-separated list of allowed origins for CORS
+# Defaults to http://localhost:3000,http://localhost:5173 when unset
+# CORS_ORIGINS=https://sparc.example.com,https://app.example.com
+
+# ---- Storage ----
+
+# Backend for patent PDF storage: "local" (default) or "s3"
+STORAGE_BACKEND=local
+
+# S3/MinIO settings (only used when STORAGE_BACKEND=s3)
+# S3_BUCKET=sparc-patents
+# S3_ENDPOINT_URL=http://localhost:9000
+# AWS_ACCESS_KEY_ID=minioadmin
+# AWS_SECRET_ACCESS_KEY=minioadmin
+# To start MinIO locally: docker compose --profile s3 up -d minio
+
+# ---- LLM ----
+
+# LLM model to use via OpenRouter
+# Supported: anthropic/claude-3.5-sonnet, openai/gpt-4o, openai/gpt-4o-mini,
+#            google/gemini-pro-1.5, meta-llama/llama-3.1-70b-instruct
+# MODEL=anthropic/claude-3.5-sonnet
+
+# ---- Cache ----
+
+# When USE_CACHE=true: check database for cached responses before making API calls
+# When USE_CACHE=false: always make fresh API calls (still stores results in database)
+USE_CACHE=true
+
+# SERP API cache TTL in hours (how long cached search results are considered fresh)
+# SERP_CACHE_TTL_HOURS=24
+
+# ---- Logging ----
+
+# Log level: DEBUG, INFO, WARNING, ERROR, CRITICAL
+# LOG_LEVEL=INFO
+
+# ---- Webhooks ----
+
+# Comma-separated list of webhook URLs for job completion and alert notifications
+# Supports generic HTTP POST and Slack/Discord incoming webhooks
+# WEBHOOK_URLS=https://hooks.slack.com/services/XXX,https://example.com/webhook

.gitea/workflows/build.yaml

@@ -0,0 +1,203 @@
+name: Build and Push Docker Images
+
+on:
+  push:
+    branches:
+      - main
+    tags:
+      - '*'
+  workflow_dispatch:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Install system dependencies
+        shell: sh
+        run: |
+          apt-get update && apt-get install -y git python3 python3-pip gcc libpq-dev python3-dev
+
+      - name: Checkout code
+        shell: sh
+        run: |
+          git clone http://gitea.gitea.svc.cluster.local/${{ gitea.repository }}.git .
+          git checkout ${{ gitea.sha }}
+
+      - name: Install Python dependencies
+        shell: sh
+        run: |
+          pip3 install -r requirements.txt ruff
+
+#      - name: Run ruff linter
+#        shell: sh
+#        run: |
+#          ruff check SPARC/ tests/
+
+      - name: Install Node.js and check TypeScript types
+        shell: sh
+        run: |
+          apt-get install -y nodejs npm
+          cd frontend
+          npm ci
+          npm run generate:local
+          if ! git diff --quiet src/api/schema.d.ts; then
+            echo "ERROR: src/api/schema.d.ts is out of date. Run 'npm run generate:local' and commit the result."
+            git diff src/api/schema.d.ts
+            exit 1
+          fi
+          npx tsc --noEmit
+
+#      - name: Run pytest
+#        shell: sh
+#        env:
+#          DATABASE_URL: "sqlite://"
+#          API_KEY: "test-key"
+#          OPENROUTER_API_KEY: "test-key"
+#          JWT_SECRET: "test-secret-for-ci"
+#          APP_ENV: "development"
+#        run: |
+#          python3 -m pytest tests/ -v --tb=short -x
+
+  build-api:
+    needs: test
+    runs-on: ubuntu-latest
+    steps:
+      - name: Install dependencies
+        shell: sh
+        run: |
+          apt-get update && apt-get install -y git docker.io
+
+      - name: Checkout code
+        shell: sh
+        run: |
+          git clone http://gitea.gitea.svc.cluster.local/${{ gitea.repository }}.git .
+          git checkout ${{ gitea.sha }}
+
+      - name: Determine image tags
+        id: tags
+        shell: sh
+        run: |
+          REGISTRY="gitea.gitea.svc.cluster.local:80"
+          REPO_OWNER="${{ gitea.repository_owner }}"
+          REPO_NAME="${{ gitea.repository }}"
+
+          REPO_NAME_ONLY=$(echo "$REPO_NAME" | cut -d'/' -f2)
+          REPO_OWNER_LOWER=$(echo "$REPO_OWNER" | tr '[:upper:]' '[:lower:]')
+          REPO_NAME_LOWER=$(echo "$REPO_NAME_ONLY" | tr '[:upper:]' '[:lower:]')
+
+          IMAGE_BASE="${REGISTRY}/${REPO_OWNER_LOWER}/${REPO_NAME_LOWER}"
+
+          case "${{ gitea.ref }}" in
+            refs/tags/*)
+              TAG_NAME="${{ gitea.ref_name }}"
+              echo "IMAGE_TAG=${IMAGE_BASE}:${TAG_NAME}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=true" >> $GITHUB_OUTPUT
+              ;;
+            refs/heads/main)
+              TIMESTAMP=$(date -u +%Y%m%d%H%M%S)
+              SHORT_SHA=$(echo "${{ gitea.sha }}" | cut -c1-7)
+              echo "IMAGE_TAG=${IMAGE_BASE}:${TIMESTAMP}-${SHORT_SHA}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=true" >> $GITHUB_OUTPUT
+              ;;
+            *)
+              BRANCH_TAG=$(echo "${{ gitea.ref_name }}" | sed 's/\//-/g')
+              echo "IMAGE_TAG=${IMAGE_BASE}:${BRANCH_TAG}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=false" >> $GITHUB_OUTPUT
+              ;;
+          esac
+
+          echo "IMAGE_LATEST=${IMAGE_BASE}:latest" >> $GITHUB_OUTPUT
+
+      - name: Login to registry
+        shell: sh
+        run: |
+          echo "${{ secrets.PERSONAL_TOKEN }}" | docker login gitea.gitea.svc.cluster.local:80 -u "${{ gitea.actor }}" --password-stdin
+
+      - name: Build and push API image
+        shell: sh
+        run: |
+          echo "Building API image..."
+          docker build -t ${{ steps.tags.outputs.IMAGE_TAG }} .
+
+          echo "Pushing API image..."
+          docker push ${{ steps.tags.outputs.IMAGE_TAG }}
+
+          if [ "${{ steps.tags.outputs.PUSH_LATEST }}" = "true" ]; then
+            echo "Tagging and pushing latest..."
+            docker tag ${{ steps.tags.outputs.IMAGE_TAG }} ${{ steps.tags.outputs.IMAGE_LATEST }}
+            docker push ${{ steps.tags.outputs.IMAGE_LATEST }}
+          fi
+
+          echo "API image available at ${{ steps.tags.outputs.IMAGE_TAG }}"
+
+  build-frontend:
+    needs: test
+    runs-on: ubuntu-latest
+    steps:
+      - name: Install dependencies
+        shell: sh
+        run: |
+           apt-get update && apt-get install -y git docker.io
+
+      - name: Checkout code
+        shell: sh
+        run: |
+          git clone http://gitea.gitea.svc.cluster.local/${{ gitea.repository }}.git .
+          git checkout ${{ gitea.sha }}
+
+      - name: Determine image tags
+        id: tags
+        shell: sh
+        run: |
+          REGISTRY="gitea.gitea.svc.cluster.local:80"
+          REPO_OWNER="${{ gitea.repository_owner }}"
+          REPO_NAME="${{ gitea.repository }}"
+
+          REPO_NAME_ONLY=$(echo "$REPO_NAME" | cut -d'/' -f2)
+          REPO_OWNER_LOWER=$(echo "$REPO_OWNER" | tr '[:upper:]' '[:lower:]')
+          REPO_NAME_LOWER=$(echo "$REPO_NAME_ONLY" | tr '[:upper:]' '[:lower:]')
+
+          IMAGE_BASE="${REGISTRY}/${REPO_OWNER_LOWER}/${REPO_NAME_LOWER}"
+
+          case "${{ gitea.ref }}" in
+            refs/tags/*)
+              TAG_NAME="${{ gitea.ref_name }}"
+              echo "IMAGE_TAG=${IMAGE_BASE}:frontend-${TAG_NAME}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=true" >> $GITHUB_OUTPUT
+              ;;
+            refs/heads/main)
+              TIMESTAMP=$(date -u +%Y%m%d%H%M%S)
+              SHORT_SHA=$(echo "${{ gitea.sha }}" | cut -c1-7)
+              echo "IMAGE_TAG=${IMAGE_BASE}:frontend-${TIMESTAMP}-${SHORT_SHA}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=true" >> $GITHUB_OUTPUT
+              ;;
+            *)
+              BRANCH_TAG=$(echo "${{ gitea.ref_name }}" | sed 's/\//-/g')
+              echo "IMAGE_TAG=${IMAGE_BASE}:frontend-${BRANCH_TAG}" >> $GITHUB_OUTPUT
+              echo "PUSH_LATEST=false" >> $GITHUB_OUTPUT
+              ;;
+          esac
+
+          echo "IMAGE_LATEST=${IMAGE_BASE}:frontend-latest" >> $GITHUB_OUTPUT
+
+      - name: Login to registry
+        shell: sh
+        run: |
+          echo "${{ secrets.PERSONAL_TOKEN }}" | docker login gitea.gitea.svc.cluster.local:80 -u "${{ gitea.actor }}" --password-stdin
+
+      - name: Build and push frontend image
+        shell: sh
+        run: |
+          echo "Building frontend image..."
+          docker build -t ${{ steps.tags.outputs.IMAGE_TAG }} ./frontend
+
+          echo "Pushing frontend image..."
+          docker push ${{ steps.tags.outputs.IMAGE_TAG }}
+
+          if [ "${{ steps.tags.outputs.PUSH_LATEST }}" = "true" ]; then
+            echo "Tagging and pushing frontend-latest..."
+            docker tag ${{ steps.tags.outputs.IMAGE_TAG }} ${{ steps.tags.outputs.IMAGE_LATEST }}
+            docker push ${{ steps.tags.outputs.IMAGE_LATEST }}
+          fi
+
+          echo "Frontend image available at ${{ steps.tags.outputs.IMAGE_TAG }}"

.gitea/workflows/test.yaml

@@ -0,0 +1,67 @@
+name: Test and Lint
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+  workflow_dispatch:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Install system dependencies
+        shell: sh
+        run: |
+          apt-get update && apt-get install -y git python3 python3-pip gcc libpq-dev python3-dev
+
+      - name: Checkout code
+        shell: sh
+        run: |
+          git clone http://gitea.gitea.svc.cluster.local/${{ gitea.repository }}.git .
+          git checkout ${{ gitea.sha }}
+
+      - name: Install Python dependencies
+        shell: sh
+        run: |
+          pip3 install -r requirements.txt ruff
+
+      - name: Run ruff linter
+        shell: sh
+        run: |
+          ruff check SPARC/ tests/
+
+      - name: Install Node.js and frontend dependencies
+        shell: sh
+        run: |
+          apt-get install -y nodejs npm
+          cd frontend && npm ci
+
+      - name: Verify generated API types are up to date
+        shell: sh
+        run: |
+          cd frontend && npm run generate:local
+          if ! git diff --quiet src/api/schema.d.ts; then
+            echo "ERROR: src/api/schema.d.ts is out of date. Run 'npm run generate:local' and commit the result."
+            git diff src/api/schema.d.ts
+            exit 1
+          fi
+
+      - name: Run TypeScript type check
+        shell: sh
+        run: |
+          cd frontend && npx tsc --noEmit
+
+      - name: Run pytest
+        shell: sh
+        env:
+          DATABASE_URL: "sqlite://"
+          API_KEY: "test-key"
+          OPENROUTER_API_KEY: "test-key"
+          JWT_SECRET: "test-secret-for-ci"
+          APP_ENV: "development"
+        run: |
+          python3 -m pytest tests/ -v --tb=short -x

.gitignore

@@ -2,4 +2,5 @@
 .pyenv
 __pycache__
 .venv
-patents
+patents
+tmp/

Dockerfile

@@ -0,0 +1,25 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+# Install system dependencies for pdfplumber and psycopg2
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    gcc \
+    libpq-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY requirements.txt .
+
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+# Create patents directory for PDF storage
+RUN mkdir -p /app/patents
+
+# Expose ports for API and Dashboard
+EXPOSE 8000 8501
+
+# Default command runs the API (can be overridden in docker-compose)
+CMD ["uvicorn", "SPARC.api:app", "--host", "0.0.0.0", "--port", "8000"]
+

README.md

@@ -1,28 +1,301 @@
 # SPARC
 
-## Name
-Semiconductor Patent & Analytics Report Core
+**Semiconductor Patent & Analytics Report Core**
 
-## Description
+A patent analysis system that estimates company performance by analyzing their patent portfolios using LLM-powered insights.
 
-## Installation
-### NixOS Installation
-`nix develop` to build and configure nix dev environment
+## Overview
 
-## Usage
-```bash
-docker compose up -d
+SPARC automatically collects, parses, and analyzes patents from companies to provide performance estimations. It uses Claude AI to evaluate innovation quality, strategic direction, and competitive positioning based on patent content.
+
+## Features
+
+- **Patent Retrieval**: Automated collection via SerpAPI's Google Patents engine
+- **Intelligent Parsing**: Extracts key sections (abstract, claims, summary) from patent PDFs
+- **Content Minimization**: Removes verbose descriptions to reduce LLM token usage
+- **AI Analysis**: Uses Claude 3.5 Sonnet via OpenRouter to analyze innovation quality and market potential
+- **Portfolio Analysis**: Evaluates multiple patents holistically for comprehensive insights
+- **Batch Processing**: Analyze multiple companies concurrently with progress tracking
+- **REST API**: FastAPI web service with async job support
+- **Dashboard**: React TypeScript web dashboard with authentication
+- **Robust Testing**: 40 tests covering all major functionality
+
+## Architecture
+
+```
+SPARC/
+├── serp_api.py       # Patent retrieval and PDF parsing
+├── llm.py            # Claude AI integration via OpenRouter
+├── analyzer.py       # High-level orchestration
+├── api.py            # FastAPI web service with auth endpoints
+├── auth.py           # JWT authentication module
+├── database.py       # PostgreSQL storage with caching
+├── types.py          # Data models
+└── config.py         # Environment configuration
 ```
 
-## Roadmap
-- [X] Retrive `publicationID` from SERP API 
-- [ ] Retrive data from Google's patent API based on those `publicationID`'s
-    - This may not be needed, looking to parse the patents based soley on the pdf retrived from SERP
-- [ ] Wrap this into a python fastAPI, then bundle with docker
+## Installation
 
+### Docker (Recommended)
+
+```bash
+# Clone and configure
+git clone <repository-url>
+cd SPARC
+cp .env.example .env
+# Edit .env with your API keys
+
+# Start all services (API, Dashboard, PostgreSQL)
+docker-compose up -d
+
+# Access the services
+# - API: http://localhost:8000
+# - Dashboard: http://localhost:8080
+# - API Docs: http://localhost:8000/docs
+```
+
+#### Patent PDF Storage
+
+The API stores downloaded patent PDFs in a `patents/` directory. In Docker,
+this is mounted as a bind mount (`./patents:/app/patents`) so that PDFs persist
+across container restarts.
+
+If you deploy to a different environment, ensure the `patents/` directory is a
+persistent volume. Without it, PDFs will be re-downloaded on every analysis.
+
+```yaml
+# docker-compose.yml excerpt
+volumes:
+  - ./patents:/app/patents
+```
+
+### NixOS
+
+```bash
+nix develop
+```
+
+This automatically creates a virtual environment and installs all dependencies.
+
+### Manual Installation
+
+```bash
+python -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+```
+
+## Configuration
+
+Create a `.env` file in the project root:
+
+```bash
+# SerpAPI key for patent search
+API_KEY=your_serpapi_key_here
+
+# OpenRouter API key for Claude AI analysis
+OPENROUTER_API_KEY=your_openrouter_key_here
+```
+
+Get your API keys:
+- SerpAPI: https://serpapi.com/
+- OpenRouter: https://openrouter.ai/
+
+## Usage
+
+### Basic Usage
+
+```python
+from SPARC.analyzer import CompanyAnalyzer
+
+# Initialize the analyzer
+analyzer = CompanyAnalyzer()
+
+# Analyze a company's patent portfolio
+analysis = analyzer.analyze_company("nvidia")
+print(analysis)
+```
+
+### Run the Example
+
+```bash
+python main.py
+```
+
+This will:
+1. Retrieve recent NVIDIA patents
+2. Parse and minimize content
+3. Analyze with Claude AI
+4. Print comprehensive performance assessment
+
+### Single Patent Analysis
+
+```python
+# Analyze a specific patent
+result = analyzer.analyze_single_patent(
+    patent_id="US11322171B1",
+    company_name="nvidia"
+)
+```
+
+### Multi-Company Batch Analysis
+
+```python
+from SPARC.analyzer import CompanyAnalyzer
+
+analyzer = CompanyAnalyzer()
+
+# Analyze multiple companies concurrently (default 3 workers)
+batch_result = analyzer.analyze_companies(
+    ["nvidia", "amd", "intel", "qualcomm"],
+    max_workers=3
+)
+
+# Access results
+print(f"Analyzed: {batch_result.total_companies}")
+print(f"Successful: {batch_result.successful}")
+print(f"Failed: {batch_result.failed}")
+
+for result in batch_result.results:
+    if result.success:
+        print(f"{result.company_name}: {result.patent_count} patents")
+        print(result.analysis)
+
+# Or use sequential processing (safer for rate limits)
+batch_result = analyzer.analyze_companies_sequential(["nvidia", "amd"])
+```
+
+### REST API
+
+Start the FastAPI server:
+
+```bash
+uvicorn SPARC.api:app --reload
+```
+
+API endpoints:
+
+| Endpoint | Method | Description |
+|----------|--------|-------------|
+| `/health` | GET | Health check |
+| `/analyze/{company}` | GET | Analyze single company |
+| `/analyze/batch` | POST | Analyze multiple companies |
+| `/analyze/batch/async` | POST | Start async batch job |
+| `/jobs/{job_id}` | GET | Get job status |
+| `/jobs` | GET | List all jobs |
+
+Interactive docs available at `http://localhost:8000/docs`
+
+Example API usage:
+
+```bash
+# Single company
+curl http://localhost:8000/analyze/nvidia
+
+# Batch analysis
+curl -X POST http://localhost:8000/analyze/batch \
+  -H "Content-Type: application/json" \
+  -d '{"companies": ["nvidia", "amd", "intel"]}'
+
+# Async batch (for long-running jobs)
+curl -X POST http://localhost:8000/analyze/batch/async \
+  -H "Content-Type: application/json" \
+  -d '{"companies": ["nvidia", "amd", "intel", "qualcomm"]}'
+```
+
+### Web Dashboard
+
+The React dashboard is included in Docker Compose:
+
+```bash
+docker-compose up -d
+```
+
+Dashboard features:
+- **Authentication**: User registration, login, and JWT-based sessions
+- **Company Analysis**: Analyze individual companies with real-time results
+- **Batch Analysis**: Process multiple companies with progress tracking
+- **Analytics**: View historical analysis data and trends
+- **Admin Panel**: User management for administrators
+
+The dashboard runs at `http://localhost:8080` when using Docker Compose.
+
+## Running Tests
+
+```bash
+# Run all tests
+pytest tests/ -v
+
+# Run specific test modules
+pytest tests/test_analyzer.py -v
+pytest tests/test_llm.py -v
+pytest tests/test_serp_api.py -v
+
+# Run with coverage
+pytest tests/ --cov=SPARC --cov-report=term-missing
+```
+
+## How It Works
+
+1. **Patent Collection**: Queries SerpAPI for company patents
+2. **PDF Download**: Retrieves patent PDF files
+3. **Section Extraction**: Parses abstract, claims, summary, and description
+4. **Content Minimization**: Keeps essential sections, removes bloated descriptions
+5. **LLM Analysis**: Sends minimized content to Claude for analysis
+6. **Performance Estimation**: Returns insights on innovation quality and outlook
+
+## Roadmap
+
+- [X] Retrieve `publicationID` from SERP API
+- [X] Parse patents from PDFs (no need for Google Patent API)
+- [X] Extract and minimize patent content
+- [X] LLM integration for analysis
+- [X] Company performance estimation
+- [X] Multi-company batch processing
+- [X] FastAPI web service wrapper
+- [X] Docker containerization
+- [X] Results persistence (database)
+- [X] Visualization dashboard
+
+## Development
+
+### Code Style
+
+- Type hints throughout
+- Comprehensive docstrings
+- Small, testable functions
+- Conventional commits
+
+### Testing Philosophy
+
+- Unit tests for core logic
+- Integration tests for orchestration
+- Mock external APIs
+- Aim for high coverage
+
+### Making Changes
+
+1. Write tests first
+2. Implement feature
+3. Verify all tests pass
+4. Commit with conventional format: `type: description`
+
+Types: `feat`, `fix`, `docs`, `test`, `refactor`, `chore`
+
+## Documentation
+
+Additional documentation is available in the `docs/` directory:
+
+- **[Deployment Guide](docs/DEPLOYMENT.md)** - Complete deployment instructions for Docker, database setup, and production configuration
+- **[Database Mode](docs/DATABASE_MODE.md)** - Database storage for prompts, responses, and analytics
+- **[Container Registry](docs/CONTAINER_REGISTRY.md)** - CI/CD and container registry setup with Gitea Actions
 
 ## License
+
 For open source projects, say how it is licensed.
 
-## Project status
-Heavy development for the limited time available to me
+## Project Status
+
+Core functionality complete. Ready for production use with API keys configured.
+
+All major features implemented: REST API, React dashboard with authentication, Docker containerization, database storage with caching, and multi-company batch processing.

ROADMAP.md

@@ -0,0 +1,122 @@
+# SPARC Roadmap
+
+Semiconductor Patent & Analytics Report Core -- development priorities.
+
+## Current State
+
+SPARC is a patent analysis platform with a working end-to-end pipeline:
+Python/FastAPI backend, React/TypeScript frontend, PostgreSQL for persistence
+and caching, Docker Compose for local development, and Gitea Actions CI/CD for
+image builds. Core features (patent retrieval via SerpAPI, PDF parsing, LLM
+analysis via OpenRouter/Claude, batch processing, JWT authentication, analytics
+dashboard) are all implemented and functional.
+
+---
+
+## P1 -- High Priority
+
+These items address correctness, security, and reliability gaps that should be
+resolved before broader production use.
+
+### Security hardening
+
+- **Rotate default JWT secret.** `auth.py` ships a fallback
+  `sparc-secret-key-change-in-production` that will be used if `JWT_SECRET` is
+  unset. Add a startup check that refuses to start with the default secret in
+  non-development environments.
+- **CORS allow-origins are hardcoded.** `api.py` only permits
+  `localhost:3000` and `localhost:5173`. Make the allowed origins configurable
+  via environment variable so the dashboard works when deployed behind a real
+  domain.
+- **Database credentials in docker-compose.yml.** The compose file embeds
+  `postgres:postgres` in plain text. Reference a `.env` file or Docker secrets
+  instead.
+
+### Error handling and resilience
+
+- **`get_db_client()` in `auth.py` creates a new `DatabaseClient` on every
+  call.** This bypasses the connection pool and can exhaust database
+  connections under load. Refactor to share a single pooled client.
+- **`_jobs` dict is in-memory only.** Job state is lost on API restart. Persist
+  job status in PostgreSQL or Redis so async batch results survive restarts.
+- **No rate limiting on auth endpoints.** `/auth/login` and `/auth/register`
+  are unprotected against brute-force or abuse. Add rate limiting middleware.
+
+### Test coverage for auth and admin
+
+- The existing API tests (`tests/test_api.py`) bypass authentication entirely.
+  Add tests that exercise the JWT flow: registration, login, protected-route
+  access, token refresh, and admin-only endpoints.
+
+---
+
+## P2 -- Medium Priority
+
+Improvements to usability, performance, and developer experience.
+
+### Backend
+
+- **Add structured logging.** Replace `print()` calls throughout `analyzer.py`,
+  `serp_api.py`, and `llm.py` with Python `logging` so log levels and
+  formatting are consistent.
+- **Make LLM model configurable.** `llm.py` hardcodes
+  `anthropic/claude-3.5-sonnet`. Accept a `MODEL` environment variable to allow
+  switching models without code changes.
+- **SERP cache TTL is hardcoded to 24 hours.** Expose `SERP_CACHE_TTL_HOURS`
+  as an environment variable in `config.py`.
+- **Patent PDF storage.** PDFs are saved to a local `patents/` directory. For
+  containerized deployments, consider object storage (S3/MinIO) or at minimum
+  document the volume mount requirement more prominently.
+- **`analyze_single_patent` assumes local file path.** The method constructs
+  `patents/{patent_id}.pdf` and reads from disk, but does not download the PDF
+  first. Either integrate the download step or document the prerequisite.
+- **`Patent.patent_id` typed as `int` in `types.py` but used as `str`
+  everywhere.** Fix the type annotation to `str`.
+
+### Frontend
+
+- **No loading/error states on several pages.** The Batch and Analytics pages
+  would benefit from skeleton loaders and user-friendly error messages.
+- **No dark mode.** Tailwind is configured but no dark variant is applied.
+- **Missing `package-lock.json` or `pnpm-lock.yaml`.** The frontend has no
+  lockfile committed, leading to non-reproducible builds.
+
+### CI/CD
+
+- **No test stage in the Gitea Actions workflow.** `build.yaml` builds and
+  pushes images but never runs `pytest`. Add a test job that gates the build.
+- **No linting or type checking.** Add `ruff` (Python) and `tsc --noEmit`
+  (TypeScript) to CI.
+
+---
+
+## P3 -- Nice to Have
+
+Lower-urgency enhancements and future features.
+
+- **Export analysis reports.** Allow users to download analysis results as PDF
+  or CSV from the dashboard.
+- **Comparison view.** Side-by-side comparison of two companies' patent
+  portfolios.
+- **Scheduled/recurring analysis.** Periodically re-analyze tracked companies
+  and alert on significant changes.
+- **Webhook/notification support.** Send alerts (Slack, Discord, email) when
+  batch jobs complete or when a company's innovation score changes
+  significantly.
+- **Multi-model support.** Let users choose between LLM providers per analysis
+  (e.g., GPT-4o, Gemini, Claude) and compare outputs.
+- **Patent trend charts.** Visualize patent filing frequency and technology
+  category distribution over time in the Analytics page.
+- **API pagination.** The `/analyze/batch` and `/jobs` endpoints could benefit
+  from cursor-based pagination for large result sets.
+- **OpenAPI client generation.** Auto-generate the TypeScript API client from
+  the FastAPI OpenAPI spec to keep frontend types in sync.
+
+---
+
+## Infrastructure and Deployment
+
+Kubernetes manifests, Helm charts, and cluster-level concerns (MetalLB,
+storage, FluxCD sync) are tracked in the
+[Talos](https://10.0.1.10/leeworks-agents/Talos) repository. File
+infrastructure-related issues there, not here.

SPARC/__init__.py

@@ -1,3 +1,4 @@
-from .types import Patents, Patent
+from .types import Patent as Patent
+from .types import Patents as Patents
 
-all = ["Patents", "Patent"]
+__all__ = ["Patents", "Patent"]

SPARC/analyzer.py

@@ -0,0 +1,361 @@
+"""High-level patent analysis orchestration.
+
+This module ties together patent retrieval, parsing, and LLM analysis
+to provide company performance estimation based on patent portfolios.
+"""
+
+import hashlib
+import logging
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from typing import Callable
+
+from SPARC import config
+
+logger = logging.getLogger(__name__)
+from SPARC.database import DatabaseClient
+from SPARC.llm import LLMAnalyzer
+from SPARC.serp_api import SERP
+from SPARC.types import BatchAnalysisResult, CompanyAnalysisResult, Patent, Patents
+
+
+class CompanyAnalyzer:
+    """Orchestrates end-to-end company performance analysis via patents."""
+
+    def __init__(self, openrouter_api_key: str | None = None, db_client: DatabaseClient | None = None):
+        """Initialize the company analyzer.
+
+        Args:
+          openrouter_api_key: Optional OpenRouter API key. If None, loads from config.
+          db_client: Optional DatabaseClient for patent caching. Created automatically if None.
+        """
+        self.llm_analyzer = LLMAnalyzer(api_key=openrouter_api_key)
+        self.db = db_client or DatabaseClient(config.database_url)
+        self.db.connect()
+        self.db.initialize_schema()
+
+    def analyze_company(self, company_name: str, patents: "Patents | None" = None, model: str | None = None) -> str:
+        """Analyze a company's performance based on their patent portfolio.
+
+        This is the main entry point that orchestrates the full pipeline:
+        1. Retrieve patents from SERP API
+        2. Download and parse each patent PDF
+        3. Minimize patent content (remove bloat)
+        4. Analyze portfolio with LLM
+        5. Return performance estimation
+
+        Args:
+          company_name: Name of the company to analyze
+          patents: Optional pre-fetched Patents result to avoid duplicate API calls
+          model: Optional LLM model override (e.g. 'openai/gpt-4o')
+
+        Returns:
+          Comprehensive analysis of company's innovation and performance outlook
+        """
+        if patents is None:
+            # Check SERP query cache first
+            query_hash = hashlib.sha256(company_name.lower().encode()).hexdigest()
+            cached_ids = self.db.get_cached_serp_query(query_hash)
+            if cached_ids is not None:
+                logger.info("Using cached SERP results for %s (%d patents)", company_name, len(cached_ids))
+                patents = Patents(patents=[
+                    Patent(patent_id=pid, pdf_link="")
+                    for pid in cached_ids
+                ])
+            else:
+                logger.info("Retrieving patents for %s...", company_name)
+                patents = SERP.query(company_name)
+                # Cache the SERP results
+                if patents.patents:
+                    self.db.store_serp_query(
+                        company_name=company_name,
+                        query_hash=query_hash,
+                        patent_ids=[p.patent_id for p in patents.patents],
+                        ttl_hours=config.serp_cache_ttl_hours,
+                    )
+
+        if not patents.patents:
+            return f"No patents found for {company_name}"
+
+        logger.info("Found %d patents. Processing...", len(patents.patents))
+
+        # Download, parse, and minimize patents in parallel
+        processed_patents = []
+        with ThreadPoolExecutor(max_workers=config.patent_thread_workers) as executor:
+            future_to_patent = {
+                executor.submit(self._process_single_patent, patent, company_name, self.db): patent
+                for patent in patents.patents
+            }
+            for future in as_completed(future_to_patent):
+                patent = future_to_patent[future]
+                try:
+                    result = future.result()
+                    if result:
+                        processed_patents.append(result)
+                except Exception as e:
+                    logger.warning("Failed to process %s: %s", patent.patent_id, e)
+
+        if not processed_patents:
+            return f"Failed to process any patents for {company_name}"
+
+        logger.info("Analyzing portfolio with LLM...")
+
+        # Analyze the full portfolio with LLM
+        analysis = self.llm_analyzer.analyze_patent_portfolio(
+            patents_data=processed_patents, company_name=company_name, model=model
+        )
+
+        return analysis
+
+    def analyze_single_patent(self, patent_id: str, company_name: str, model: str | None = None) -> str:
+        """Analyze a single patent by ID.
+
+        If the patent PDF is not already on disk, this method attempts to
+        download it automatically by looking up the PDF link in the database
+        cache. If the link is not cached either, a ``FileNotFoundError`` is
+        raised with instructions on how to obtain the PDF.
+
+        Args:
+          patent_id: Publication ID of the patent (e.g. "US-11234567-B2")
+          company_name: Name of the company (for context)
+          model: Optional LLM model override (e.g. 'openai/gpt-4o')
+
+        Returns:
+          Analysis of the specific patent's innovation quality
+
+        Raises:
+          FileNotFoundError: If the patent PDF cannot be found or downloaded.
+        """
+        import os
+        logger.info("Analyzing patent %s for %s...", patent_id, company_name)
+
+        patent_path = f"patents/{patent_id}.pdf"
+
+        if not os.path.exists(patent_path):
+            # Attempt to download the PDF automatically from cached metadata
+            cached = self.db.get_cached_patent(patent_id)
+            pdf_link = cached.get("pdf_link") if cached else None
+
+            if pdf_link:
+                logger.info("PDF not on disk; downloading %s from cached link", patent_id)
+                patent = SERP.save_patents(
+                    Patent(patent_id=patent_id, pdf_link=pdf_link)
+                )
+                patent_path = patent.pdf_path
+            else:
+                raise FileNotFoundError(
+                    f"Patent PDF not found at '{patent_path}' and no download link is "
+                    f"cached for '{patent_id}'. Run a company analysis first to populate "
+                    f"the cache, or call SERP.save_patents() with the patent's PDF link."
+                )
+
+        try:
+            sections = SERP.parse_patent_pdf(patent_path)
+            minimized_content = SERP.minimize_patent_for_llm(sections)
+
+            analysis = self.llm_analyzer.analyze_patent_content(
+                patent_content=minimized_content, company_name=company_name, model=model
+            )
+
+            return analysis
+
+        except FileNotFoundError:
+            raise
+        except Exception as e:
+            return f"Failed to analyze patent {patent_id}: {e}"
+
+    @staticmethod
+    def _process_single_patent(
+        patent: Patent,
+        company_name: str = "",
+        db: DatabaseClient | None = None,
+    ) -> dict | None:
+        """Download, parse, and minimize a single patent. Thread-safe.
+
+        Checks DB cache before downloading. Stores results after processing.
+
+        Returns:
+            Dict with patent_id and minimized content, or None on failure.
+        """
+        try:
+            # Check DB cache first
+            if db:
+                cached = db.get_cached_patent(patent.patent_id)
+                if cached and cached.get("minimized_content"):
+                    return {"patent_id": patent.patent_id, "content": cached["minimized_content"]}
+
+            # Full processing: download, parse, minimize
+            patent = SERP.save_patents(patent)
+            sections = SERP.parse_patent_pdf(patent.pdf_path)
+            minimized_content = SERP.minimize_patent_for_llm(sections)
+
+            # Store in DB cache
+            if db:
+                db.store_patent(
+                    patent_id=patent.patent_id,
+                    company_name=company_name,
+                    pdf_link=patent.pdf_link,
+                    raw_sections=sections,
+                    minimized_content=minimized_content,
+                )
+
+            return {"patent_id": patent.patent_id, "content": minimized_content}
+        except Exception as e:
+            logger.warning("Failed to process %s: %s", patent.patent_id, e)
+            return None
+
+    def _analyze_company_safe(self, company_name: str, model: str | None = None) -> CompanyAnalysisResult:
+        """Internal wrapper that catches exceptions and returns structured result.
+
+        Args:
+            company_name: Name of the company to analyze
+            model: Optional LLM model override (e.g. 'openai/gpt-4o')
+
+        Returns:
+            CompanyAnalysisResult with success/failure status
+        """
+        try:
+            # Delegate to analyze_company which handles SERP/patent caching
+            analysis = self.analyze_company(company_name, model=model)
+
+            # Determine patent count from cached SERP query
+            query_hash = hashlib.sha256(company_name.lower().encode()).hexdigest()
+            cached_ids = self.db.get_cached_serp_query(query_hash)
+            patent_count = len(cached_ids) if cached_ids else 0
+
+            # Check if analysis indicates failure
+            if analysis.startswith("No patents found") or analysis.startswith(
+                "Failed to process"
+            ):
+                return CompanyAnalysisResult(
+                    company_name=company_name,
+                    analysis=analysis,
+                    patent_count=patent_count,
+                    success=False,
+                    error=analysis,
+                )
+
+            return CompanyAnalysisResult(
+                company_name=company_name,
+                analysis=analysis,
+                patent_count=patent_count,
+                success=True,
+            )
+
+        except Exception as e:
+            return CompanyAnalysisResult(
+                company_name=company_name,
+                analysis="",
+                patent_count=0,
+                success=False,
+                error=str(e),
+            )
+
+    def analyze_companies(
+        self,
+        companies: list[str],
+        max_workers: int = 3,
+        progress_callback: Callable[[str, int, int], None] | None = None,
+        model: str | None = None,
+    ) -> BatchAnalysisResult:
+        """Analyze multiple companies' patent portfolios in batch.
+
+        Processes companies concurrently for improved performance while
+        respecting API rate limits.
+
+        Args:
+            companies: List of company names to analyze
+            max_workers: Maximum concurrent analyses (default 3 to avoid rate limits)
+            progress_callback: Optional callback(company_name, completed, total)
+            model: Optional LLM model override (e.g. 'openai/gpt-4o')
+
+        Returns:
+            BatchAnalysisResult containing all individual results and summary stats
+        """
+        results: list[CompanyAnalysisResult] = []
+        total = len(companies)
+
+        logger.info("Starting batch analysis of %d companies...", total)
+
+        with ThreadPoolExecutor(max_workers=max_workers) as executor:
+            future_to_company = {
+                executor.submit(self._analyze_company_safe, company, model): company
+                for company in companies
+            }
+
+            completed = 0
+            for future in as_completed(future_to_company):
+                company = future_to_company[future]
+                completed += 1
+
+                try:
+                    result = future.result()
+                    results.append(result)
+
+                    status = "OK" if result.success else "FAIL"
+                    logger.info("[%d/%d] %s %s", completed, total, status, company)
+
+                    if progress_callback:
+                        progress_callback(company, completed, total)
+
+                except Exception as e:
+                    results.append(
+                        CompanyAnalysisResult(
+                            company_name=company,
+                            analysis="",
+                            patent_count=0,
+                            success=False,
+                            error=str(e),
+                        )
+                    )
+                    logger.error("[%d/%d] FAIL %s: %s", completed, total, company, e)
+
+        successful = sum(1 for r in results if r.success)
+        failed = total - successful
+
+        logger.info("Batch complete: %d succeeded, %d failed", successful, failed)
+
+        return BatchAnalysisResult(
+            results=results,
+            total_companies=total,
+            successful=successful,
+            failed=failed,
+        )
+
+    def analyze_companies_sequential(
+        self, companies: list[str]
+    ) -> BatchAnalysisResult:
+        """Analyze multiple companies sequentially (safer for rate limits).
+
+        Use this when you want more control over API rate limiting or
+        when debugging issues.
+
+        Args:
+            companies: List of company names to analyze
+
+        Returns:
+            BatchAnalysisResult containing all individual results
+        """
+        results: list[CompanyAnalysisResult] = []
+        total = len(companies)
+
+        logger.info("Starting sequential analysis of %d companies...", total)
+
+        for idx, company in enumerate(companies, 1):
+            logger.info("[%d/%d] Analyzing %s...", idx, total, company)
+            result = self._analyze_company_safe(company)
+            results.append(result)
+
+            status = "OK" if result.success else "FAIL"
+            logger.info("[%d/%d] %s %s", idx, total, status, company)
+
+        successful = sum(1 for r in results if r.success)
+        failed = total - successful
+
+        logger.info("Batch complete: %d succeeded, %d failed", successful, failed)
+
+        return BatchAnalysisResult(
+            results=results,
+            total_companies=total,
+            successful=successful,
+            failed=failed,
+        )

SPARC/auth.py

@@ -0,0 +1,249 @@
+"""JWT authentication utilities for SPARC API."""
+
+import os
+from datetime import datetime, timedelta, timezone
+from typing import Optional
+
+import jwt
+from fastapi import Depends, HTTPException, status
+from fastapi.security import HTTPAuthorizationCredentials, HTTPBearer
+from pydantic import BaseModel
+
+from SPARC import config
+from SPARC.database import DatabaseClient
+
+# JWT Configuration
+_DEFAULT_JWT_SECRET = "sparc-secret-key-change-in-production"
+JWT_SECRET = os.getenv("JWT_SECRET", _DEFAULT_JWT_SECRET)
+JWT_ALGORITHM = "HS256"
+ACCESS_TOKEN_EXPIRE_MINUTES = 30
+REFRESH_TOKEN_EXPIRE_DAYS = 7
+
+
+def check_jwt_secret() -> None:
+    """Refuse to start with the default JWT secret in non-development environments.
+
+    Raises:
+        RuntimeError: If JWT_SECRET is the default value and APP_ENV is not 'development'.
+    """
+    if JWT_SECRET == _DEFAULT_JWT_SECRET and config.app_env != "development":
+        raise RuntimeError(
+            f"FATAL: JWT_SECRET is set to the default value and APP_ENV={config.app_env!r}. "
+            "Set a secure JWT_SECRET environment variable before running in non-development environments."
+        )
+
+security = HTTPBearer()
+
+
+class TokenPayload(BaseModel):
+    """JWT token payload."""
+
+    sub: str  # user_id as string (JWT RFC 7519 requires sub to be a string)
+    email: str
+    role: str
+    exp: datetime
+    type: str  # "access" or "refresh"
+
+    @property
+    def user_id(self) -> int:
+        """Get user_id as integer."""
+        return int(self.sub)
+
+
+class TokenResponse(BaseModel):
+    """Token response model."""
+
+    access_token: str
+    refresh_token: str
+    token_type: str = "bearer"
+
+
+class UserResponse(BaseModel):
+    """User response model."""
+
+    id: int
+    email: str
+    role: str
+    created_at: datetime
+
+
+def create_access_token(user_id: int, email: str, role: str) -> str:
+    """Create a JWT access token.
+
+    Args:
+        user_id: User ID
+        email: User email
+        role: User role
+
+    Returns:
+        Encoded JWT token
+    """
+    expire = datetime.now(timezone.utc) + timedelta(minutes=ACCESS_TOKEN_EXPIRE_MINUTES)
+    payload = {
+        "sub": str(user_id),
+        "email": email,
+        "role": role,
+        "exp": expire,
+        "type": "access",
+    }
+    return jwt.encode(payload, JWT_SECRET, algorithm=JWT_ALGORITHM)
+
+
+def create_refresh_token(user_id: int, email: str, role: str) -> str:
+    """Create a JWT refresh token.
+
+    Args:
+        user_id: User ID
+        email: User email
+        role: User role
+
+    Returns:
+        Encoded JWT token
+    """
+    expire = datetime.now(timezone.utc) + timedelta(days=REFRESH_TOKEN_EXPIRE_DAYS)
+    payload = {
+        "sub": str(user_id),
+        "email": email,
+        "role": role,
+        "exp": expire,
+        "type": "refresh",
+    }
+    return jwt.encode(payload, JWT_SECRET, algorithm=JWT_ALGORITHM)
+
+
+def create_tokens(user_id: int, email: str, role: str) -> TokenResponse:
+    """Create both access and refresh tokens.
+
+    Args:
+        user_id: User ID
+        email: User email
+        role: User role
+
+    Returns:
+        TokenResponse with both tokens
+    """
+    return TokenResponse(
+        access_token=create_access_token(user_id, email, role),
+        refresh_token=create_refresh_token(user_id, email, role),
+    )
+
+
+def decode_token(token: str) -> Optional[TokenPayload]:
+    """Decode and validate a JWT token.
+
+    Args:
+        token: JWT token string
+
+    Returns:
+        TokenPayload if valid, None otherwise
+    """
+    try:
+        payload = jwt.decode(token, JWT_SECRET, algorithms=[JWT_ALGORITHM])
+        return TokenPayload(**payload)
+    except jwt.ExpiredSignatureError:
+        return None
+    except jwt.InvalidTokenError:
+        return None
+
+
+# Shared database client singleton, initialized at startup via init_db_client()
+_db_client: DatabaseClient | None = None
+
+
+def init_db_client() -> None:
+    """Initialize the shared database client. Call once at app startup."""
+    global _db_client
+    _db_client = DatabaseClient(config.database_url)
+    _db_client.connect()
+
+
+def close_db_client() -> None:
+    """Close the shared database client. Call at app shutdown."""
+    global _db_client
+    if _db_client:
+        _db_client.close()
+        _db_client = None
+
+
+def get_db_client() -> DatabaseClient:
+    """Get the shared pooled database client for auth operations.
+
+    Returns the module-level singleton DatabaseClient. If not yet initialized
+    (e.g., during tests), creates a new instance as a fallback.
+    """
+    global _db_client
+    if _db_client is None:
+        _db_client = DatabaseClient(config.database_url)
+        _db_client.connect()
+    return _db_client
+
+
+async def get_current_user(
+    credentials: HTTPAuthorizationCredentials = Depends(security),
+) -> UserResponse:
+    """Get the current authenticated user from JWT token.
+
+    Args:
+        credentials: Bearer token from request
+
+    Returns:
+        UserResponse with user details
+
+    Raises:
+        HTTPException: If token is invalid or expired
+    """
+    token = credentials.credentials
+    payload = decode_token(token)
+
+    if not payload:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="Invalid or expired token",
+            headers={"WWW-Authenticate": "Bearer"},
+        )
+
+    if payload.type != "access":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="Invalid token type",
+            headers={"WWW-Authenticate": "Bearer"},
+        )
+
+    db = get_db_client()
+    user = db.get_user_by_id(payload.user_id)
+
+    if not user:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="User not found",
+            headers={"WWW-Authenticate": "Bearer"},
+        )
+
+    return UserResponse(
+        id=user["id"],
+        email=user["email"],
+        role=user["role"],
+        created_at=user["created_at"],
+    )
+
+
+async def get_current_admin(
+    current_user: UserResponse = Depends(get_current_user),
+) -> UserResponse:
+    """Require admin role for the current user.
+
+    Args:
+        current_user: Current authenticated user
+
+    Returns:
+        UserResponse if admin
+
+    Raises:
+        HTTPException: If user is not admin
+    """
+    if current_user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_403_FORBIDDEN,
+            detail="Admin access required",
+        )
+    return current_user

SPARC/config.py

@@ -2,13 +2,69 @@
 
 Loads environment variables from .env file for API keys and other secrets.
 """
-from dotenv import load_dotenv
+import logging
 import os
 
+from dotenv import load_dotenv
+
 load_dotenv()
 
+# Logging configuration
+log_level = os.getenv("LOG_LEVEL", "INFO").upper()
+logging.basicConfig(
+    level=getattr(logging, log_level, logging.INFO),
+    format="%(asctime)s %(levelname)s %(name)s %(message)s",
+)
+
 # SerpAPI key for patent search
 api_key = os.getenv("API_KEY")
 
-# Anthropic API key for LLM analysis
-anthropic_api_key = os.getenv("ANTHROPIC_API_KEY")
+# OpenRouter API key for LLM analysis
+openrouter_api_key = os.getenv("OPENROUTER_API_KEY")
+
+# Database configuration - all messages are stored in the database
+# The database serves as both a persistent store and a cache layer
+database_url = os.getenv("DATABASE_URL", "postgresql://postgres:postgres@localhost:5432/sparc")
+
+# Cache configuration
+# When enabled (default), the system checks the database for cached responses
+# before making API calls, saving tokens and reducing latency
+use_cache = os.getenv("USE_CACHE", "true").lower() in ("true", "1", "yes")
+
+# Legacy compatibility - USE_DATABASE is deprecated, database is always used
+# This variable is kept for backwards compatibility but has no effect
+use_database = os.getenv("USE_DATABASE", "false").lower() in ("true", "1", "yes")
+
+# Patent search configuration
+patent_search_days = int(os.getenv("PATENT_SEARCH_DAYS", "90"))
+patent_thread_workers = int(os.getenv("PATENT_THREAD_WORKERS", "5"))
+
+# LLM model to use via OpenRouter (e.g. "anthropic/claude-3.5-sonnet", "openai/gpt-4o")
+model = os.getenv("MODEL", "anthropic/claude-3.5-sonnet")
+
+# SERP cache TTL in hours (how long cached search results are considered fresh)
+serp_cache_ttl_hours = int(os.getenv("SERP_CACHE_TTL_HOURS", "24"))
+
+# Root path for running behind a reverse proxy (e.g., "/api" when served at /api/)
+# This ensures OpenAPI docs work correctly when accessed via the proxy
+root_path = os.getenv("ROOT_PATH", "")
+
+# Application environment: "development", "staging", or "production"
+# Used for safety checks (e.g., refusing default JWT secret in production)
+app_env = os.getenv("APP_ENV", "development")
+
+# Storage backend: "local" (default) or "s3" for S3/MinIO object storage
+storage_backend = os.getenv("STORAGE_BACKEND", "local")
+s3_bucket = os.getenv("S3_BUCKET", "sparc-patents")
+s3_endpoint_url = os.getenv("S3_ENDPOINT_URL", "")
+s3_access_key = os.getenv("AWS_ACCESS_KEY_ID", "")
+s3_secret_key = os.getenv("AWS_SECRET_ACCESS_KEY", "")
+
+# CORS allowed origins (comma-separated)
+# Defaults to localhost dev origins when unset
+_cors_origins_raw = os.getenv("CORS_ORIGINS", "")
+cors_origins: list[str] = (
+    [o.strip() for o in _cors_origins_raw.split(",") if o.strip()]
+    if _cors_origins_raw
+    else ["http://localhost:3000", "http://localhost:5173"]
+)

SPARC/database.py

@@ -0,0 +1,937 @@
+"""Database client for storing and retrieving LLM messages and user authentication."""
+
+import contextlib
+import hashlib
+import json
+from datetime import datetime, timedelta
+from typing import Dict, List, Optional
+
+import bcrypt
+import psycopg2
+from psycopg2.extras import RealDictCursor
+from psycopg2.pool import ThreadedConnectionPool
+
+
+class DatabaseClient:
+    """Handles database operations for message storage and retrieval."""
+
+    def __init__(self, database_url: str, minconn: int = 2, maxconn: int = 10):
+        """Initialize the database client.
+
+        Args:
+            database_url: PostgreSQL connection string
+            minconn: Minimum connections in the pool
+            maxconn: Maximum connections in the pool
+        """
+        self.database_url = database_url
+        self._pool: ThreadedConnectionPool | None = None
+        self._minconn = minconn
+        self._maxconn = maxconn
+        # Legacy single connection kept for backwards compatibility
+        self.conn = None
+
+    def _ensure_pool(self):
+        """Create the connection pool if it doesn't exist yet."""
+        if self._pool is None or self._pool.closed:
+            self._pool = ThreadedConnectionPool(
+                self._minconn, self._maxconn, self.database_url
+            )
+
+    @contextlib.contextmanager
+    def get_conn(self):
+        """Check out a connection from the pool. Returns it on exit."""
+        self._ensure_pool()
+        conn = self._pool.getconn()
+        try:
+            yield conn
+        finally:
+            self._pool.putconn(conn)
+
+    def connect(self):
+        """Establish database connection (legacy single-connection path)."""
+        if not self.conn or self.conn.closed:
+            self.conn = psycopg2.connect(self.database_url)
+
+    def close(self):
+        """Close database connection and pool."""
+        if self.conn and not self.conn.closed:
+            self.conn.close()
+        if self._pool and not self._pool.closed:
+            self._pool.closeall()
+
+    def initialize_schema(self):
+        """Create database tables if they don't exist."""
+        self.connect()
+
+        with self.conn.cursor() as cursor:
+            # Create messages table
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS llm_messages (
+                    id SERIAL PRIMARY KEY,
+                    timestamp TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    company_name VARCHAR(255),
+                    analysis_type VARCHAR(50),
+                    model VARCHAR(100),
+                    prompt TEXT NOT NULL,
+                    prompt_hash VARCHAR(64),
+                    response TEXT,
+                    metadata JSONB,
+                    token_usage JSONB,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    is_cached BOOLEAN DEFAULT FALSE
+                )
+            """)
+
+            # Create index on timestamp for analytics queries
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_messages_timestamp
+                ON llm_messages(timestamp)
+            """)
+
+            # Create index on company_name for filtering
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_messages_company
+                ON llm_messages(company_name)
+            """)
+
+            # Add prompt_hash and is_cached columns if they don't exist (for existing tables)
+            # This must run BEFORE creating the index on prompt_hash
+            cursor.execute("""
+                DO $$
+                BEGIN
+                    IF NOT EXISTS (
+                        SELECT 1 FROM information_schema.columns
+                        WHERE table_name = 'llm_messages' AND column_name = 'prompt_hash'
+                    ) THEN
+                        ALTER TABLE llm_messages ADD COLUMN prompt_hash VARCHAR(64);
+                    END IF;
+                    IF NOT EXISTS (
+                        SELECT 1 FROM information_schema.columns
+                        WHERE table_name = 'llm_messages' AND column_name = 'is_cached'
+                    ) THEN
+                        ALTER TABLE llm_messages ADD COLUMN is_cached BOOLEAN DEFAULT FALSE;
+                    END IF;
+                END $$;
+            """)
+
+            # Create index on prompt_hash for cache lookups
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_messages_prompt_hash
+                ON llm_messages(prompt_hash)
+            """)
+
+            # Create users table for authentication
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS users (
+                    id SERIAL PRIMARY KEY,
+                    email VARCHAR(255) UNIQUE NOT NULL,
+                    password_hash VARCHAR(255) NOT NULL,
+                    role VARCHAR(20) DEFAULT 'user' CHECK (role IN ('admin', 'user')),
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            # Create index on email for fast lookups
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_users_email
+                ON users(email)
+            """)
+
+            # Create patents cache table
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS patents (
+                    patent_id VARCHAR(64) PRIMARY KEY,
+                    company_name VARCHAR(255),
+                    pdf_link TEXT,
+                    raw_sections JSONB,
+                    minimized_content TEXT,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_patents_company
+                ON patents(company_name)
+            """)
+
+            # Create SERP query cache table
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS serp_queries (
+                    id SERIAL PRIMARY KEY,
+                    company_name VARCHAR(255),
+                    query_hash VARCHAR(64) UNIQUE,
+                    result_patent_ids TEXT[],
+                    expires_at TIMESTAMP NOT NULL,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_serp_queries_hash
+                ON serp_queries(query_hash)
+            """)
+
+            # Create jobs table for persisting async batch job state
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS jobs (
+                    job_id VARCHAR(128) PRIMARY KEY,
+                    status VARCHAR(20) NOT NULL DEFAULT 'pending',
+                    progress INTEGER NOT NULL DEFAULT 0,
+                    total_companies INTEGER NOT NULL DEFAULT 0,
+                    completed_companies INTEGER NOT NULL DEFAULT 0,
+                    result_json JSONB,
+                    error TEXT,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_jobs_status
+                ON jobs(status)
+            """)
+
+            # Create tracked companies table for scheduled analysis
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS tracked_companies (
+                    id SERIAL PRIMARY KEY,
+                    company_name VARCHAR(255) UNIQUE NOT NULL,
+                    last_patent_count INTEGER DEFAULT 0,
+                    last_analysis_at TIMESTAMP,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            # Create alerts table for significant changes
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS alerts (
+                    id SERIAL PRIMARY KEY,
+                    company_name VARCHAR(255) NOT NULL,
+                    alert_type VARCHAR(50) NOT NULL,
+                    message TEXT NOT NULL,
+                    old_value NUMERIC,
+                    new_value NUMERIC,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+                )
+            """)
+
+            cursor.execute("""
+                CREATE INDEX IF NOT EXISTS idx_alerts_company
+                ON alerts(company_name)
+            """)
+
+            self.conn.commit()
+
+    @staticmethod
+    def hash_prompt(prompt: str) -> str:
+        """Generate a hash of the prompt for cache lookups.
+
+        Args:
+            prompt: The prompt text to hash
+
+        Returns:
+            SHA-256 hash of the prompt
+        """
+        return hashlib.sha256(prompt.encode()).hexdigest()
+
+    def get_cached_response(
+        self,
+        prompt: str,
+        company_name: Optional[str] = None,
+        analysis_type: Optional[str] = None,
+    ) -> Optional[Dict]:
+        """Look up a cached response for a given prompt.
+
+        Args:
+            prompt: The prompt to look up
+            company_name: Optional company name filter
+            analysis_type: Optional analysis type filter
+
+        Returns:
+            Cached message dict if found, None otherwise
+        """
+        prompt_hash = self.hash_prompt(prompt)
+
+        query = """
+            SELECT * FROM llm_messages
+            WHERE prompt_hash = %s
+            AND response IS NOT NULL
+            AND response NOT LIKE '[DATABASE MODE]%%'
+            AND response NOT LIKE '[TEST MODE]%%'
+            AND response NOT LIKE '[NO API]%%'
+        """
+        params = [prompt_hash]
+
+        if company_name:
+            query += " AND company_name = %s"
+            params.append(company_name)
+
+        if analysis_type:
+            query += " AND analysis_type = %s"
+            params.append(analysis_type)
+
+        query += " ORDER BY timestamp DESC LIMIT 1"
+
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(query, params)
+                result = cursor.fetchone()
+                return dict(result) if result else None
+
+    def store_message(
+        self,
+        prompt: str,
+        response: str,
+        company_name: Optional[str] = None,
+        analysis_type: Optional[str] = None,
+        model: Optional[str] = None,
+        metadata: Optional[Dict] = None,
+        token_usage: Optional[Dict] = None,
+        is_cached: bool = False,
+    ) -> int:
+        """Store an LLM message exchange in the database.
+
+        Args:
+            prompt: The prompt sent to the LLM
+            response: The response from the LLM
+            company_name: Name of company being analyzed
+            analysis_type: Type of analysis (e.g., 'single_patent', 'portfolio')
+            model: Model identifier used
+            metadata: Additional metadata as dict
+            token_usage: Token usage information
+            is_cached: Whether this response was served from cache
+
+        Returns:
+            The ID of the inserted record
+        """
+        prompt_hash = self.hash_prompt(prompt)
+
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """
+                    INSERT INTO llm_messages
+                    (prompt, prompt_hash, response, company_name, analysis_type, model, metadata, token_usage, is_cached)
+                    VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s)
+                    RETURNING id
+                    """,
+                    (
+                        prompt,
+                        prompt_hash,
+                        response,
+                        company_name,
+                        analysis_type,
+                        model,
+                        json.dumps(metadata) if metadata else None,
+                        json.dumps(token_usage) if token_usage else None,
+                        is_cached,
+                    ),
+                )
+
+                message_id = cursor.fetchone()[0]
+            conn.commit()
+
+            return message_id
+
+    def get_messages(
+        self,
+        company_name: Optional[str] = None,
+        analysis_type: Optional[str] = None,
+        limit: int = 100,
+        offset: int = 0,
+    ) -> List[Dict]:
+        """Retrieve messages from the database.
+
+        Args:
+            company_name: Filter by company name
+            analysis_type: Filter by analysis type
+            limit: Maximum number of records to return
+            offset: Number of records to skip
+
+        Returns:
+            List of message dictionaries
+        """
+        query = "SELECT * FROM llm_messages WHERE 1=1"
+        params = []
+
+        if company_name:
+            query += " AND company_name = %s"
+            params.append(company_name)
+
+        if analysis_type:
+            query += " AND analysis_type = %s"
+            params.append(analysis_type)
+
+        query += " ORDER BY timestamp DESC LIMIT %s OFFSET %s"
+        params.extend([limit, offset])
+
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(query, params)
+                return [dict(row) for row in cursor.fetchall()]
+
+    def get_analytics(self, days: int = 30) -> Dict:
+        """Get analytics on message usage.
+
+        Args:
+            days: Number of days to look back
+
+        Returns:
+            Dictionary with analytics data
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                # Total messages
+                cursor.execute(
+                    """
+                    SELECT COUNT(*) as total_messages
+                    FROM llm_messages
+                    WHERE timestamp >= NOW() - INTERVAL '%s days'
+                    """,
+                    (days,),
+                )
+                total = cursor.fetchone()["total_messages"]
+
+                # Messages by company
+                cursor.execute(
+                    """
+                    SELECT company_name, COUNT(*) as count
+                    FROM llm_messages
+                    WHERE timestamp >= NOW() - INTERVAL '%s days'
+                    GROUP BY company_name
+                    ORDER BY count DESC
+                    LIMIT 10
+                    """,
+                    (days,),
+                )
+                by_company = cursor.fetchall()
+
+                # Messages by type
+                cursor.execute(
+                    """
+                    SELECT analysis_type, COUNT(*) as count
+                    FROM llm_messages
+                    WHERE timestamp >= NOW() - INTERVAL '%s days'
+                    GROUP BY analysis_type
+                    ORDER BY count DESC
+                    """,
+                    (days,),
+                )
+                by_type = cursor.fetchall()
+
+                return {
+                    "total_messages": total,
+                    "by_company": [dict(row) for row in by_company],
+                    "by_type": [dict(row) for row in by_type],
+                    "period_days": days,
+                }
+
+    # Patent Cache Methods
+
+    def get_cached_patent(self, patent_id: str) -> Optional[Dict]:
+        """Look up a cached patent by ID.
+
+        Returns:
+            Dict with raw_sections and minimized_content, or None.
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    "SELECT * FROM patents WHERE patent_id = %s",
+                    (patent_id,),
+                )
+                row = cursor.fetchone()
+                return dict(row) if row else None
+
+    def store_patent(
+        self,
+        patent_id: str,
+        company_name: str,
+        pdf_link: str,
+        raw_sections: Dict,
+        minimized_content: str,
+    ) -> None:
+        """Store a processed patent in the cache."""
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """
+                    INSERT INTO patents (patent_id, company_name, pdf_link, raw_sections, minimized_content)
+                    VALUES (%s, %s, %s, %s, %s)
+                    ON CONFLICT (patent_id) DO UPDATE SET
+                        raw_sections = EXCLUDED.raw_sections,
+                        minimized_content = EXCLUDED.minimized_content
+                    """,
+                    (patent_id, company_name, pdf_link, json.dumps(raw_sections), minimized_content),
+                )
+            conn.commit()
+
+    def get_cached_serp_query(self, query_hash: str) -> Optional[List[str]]:
+        """Look up cached SERP query results.
+
+        Returns:
+            List of patent IDs if cache hit and not expired, None otherwise.
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    """
+                    SELECT result_patent_ids FROM serp_queries
+                    WHERE query_hash = %s AND expires_at > NOW()
+                    """,
+                    (query_hash,),
+                )
+                row = cursor.fetchone()
+                return row["result_patent_ids"] if row else None
+
+    def store_serp_query(
+        self,
+        company_name: str,
+        query_hash: str,
+        patent_ids: List[str],
+        ttl_hours: int = 24,
+    ) -> None:
+        """Store SERP query results in the cache."""
+        expires_at = datetime.now() + timedelta(hours=ttl_hours)
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """
+                    INSERT INTO serp_queries (company_name, query_hash, result_patent_ids, expires_at)
+                    VALUES (%s, %s, %s, %s)
+                    ON CONFLICT (query_hash) DO UPDATE SET
+                        result_patent_ids = EXCLUDED.result_patent_ids,
+                        expires_at = EXCLUDED.expires_at
+                    """,
+                    (company_name, query_hash, patent_ids, expires_at),
+                )
+            conn.commit()
+
+    # Job Persistence Methods
+
+    def create_job(
+        self,
+        job_id: str,
+        total_companies: int,
+    ) -> Dict:
+        """Create a new job record.
+
+        Args:
+            job_id: Unique job identifier
+            total_companies: Number of companies in the batch
+
+        Returns:
+            Job dict
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    """
+                    INSERT INTO jobs (job_id, status, progress, total_companies, completed_companies)
+                    VALUES (%s, 'pending', 0, %s, 0)
+                    RETURNING *
+                    """,
+                    (job_id, total_companies),
+                )
+                job = cursor.fetchone()
+            conn.commit()
+            return dict(job)
+
+    def update_job(
+        self,
+        job_id: str,
+        status: Optional[str] = None,
+        progress: Optional[int] = None,
+        completed_companies: Optional[int] = None,
+        result_json: Optional[str] = None,
+        error: Optional[str] = None,
+    ) -> Optional[Dict]:
+        """Update a job's state.
+
+        Only non-None fields are updated.
+        """
+        updates = []
+        params = []
+        if status is not None:
+            updates.append("status = %s")
+            params.append(status)
+        if progress is not None:
+            updates.append("progress = %s")
+            params.append(progress)
+        if completed_companies is not None:
+            updates.append("completed_companies = %s")
+            params.append(completed_companies)
+        if result_json is not None:
+            updates.append("result_json = %s")
+            params.append(result_json)
+        if error is not None:
+            updates.append("error = %s")
+            params.append(error)
+
+        if not updates:
+            return self.get_job(job_id)
+
+        updates.append("updated_at = CURRENT_TIMESTAMP")
+        params.append(job_id)
+
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    f"UPDATE jobs SET {', '.join(updates)} WHERE job_id = %s RETURNING *",
+                    params,
+                )
+                job = cursor.fetchone()
+            conn.commit()
+            return dict(job) if job else None
+
+    def get_job(self, job_id: str) -> Optional[Dict]:
+        """Get a job by ID."""
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute("SELECT * FROM jobs WHERE job_id = %s", (job_id,))
+                job = cursor.fetchone()
+                return dict(job) if job else None
+
+    def list_jobs(
+        self,
+        status: Optional[str] = None,
+        limit: int = 10,
+        cursor: Optional[str] = None,
+    ) -> List[Dict]:
+        """List jobs with optional status filter and cursor-based pagination.
+
+        Args:
+            status: Optional status filter (pending, running, completed, failed).
+            limit: Maximum number of jobs to return.
+            cursor: Opaque cursor (``created_at|job_id``) from a previous
+                response. When provided, only jobs older than the cursor are
+                returned.
+
+        Returns:
+            List of job dicts ordered by created_at descending.
+        """
+        conditions: list[str] = []
+        params: list = []
+
+        if status:
+            conditions.append("status = %s")
+            params.append(status)
+
+        if cursor:
+            try:
+                ts_str, cursor_job_id = cursor.rsplit("|", 1)
+                conditions.append("(created_at, job_id) < (%s, %s)")
+                params.extend([ts_str, cursor_job_id])
+            except ValueError:
+                pass  # Ignore malformed cursors; return from start
+
+        query = "SELECT * FROM jobs"
+        if conditions:
+            query += " WHERE " + " AND ".join(conditions)
+        query += " ORDER BY created_at DESC, job_id DESC LIMIT %s"
+        params.append(limit)
+
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cur:
+                cur.execute(query, params)
+                return [dict(row) for row in cur.fetchall()]
+
+    def mark_stale_jobs_failed(self) -> int:
+        """Mark any jobs in 'running' or 'pending' state as 'failed'.
+
+        Called at startup to clean up jobs that were interrupted by a restart.
+
+        Returns:
+            Number of jobs marked as failed.
+        """
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """
+                    UPDATE jobs SET status = 'failed', error = 'Interrupted by server restart',
+                    updated_at = CURRENT_TIMESTAMP
+                    WHERE status IN ('running', 'pending')
+                    """
+                )
+                count = cursor.rowcount
+            conn.commit()
+            return count
+
+    # User Authentication Methods
+
+    @staticmethod
+    def hash_password(password: str) -> str:
+        """Hash a password using bcrypt.
+
+        Args:
+            password: Plain text password
+
+        Returns:
+            Hashed password string
+        """
+        return bcrypt.hashpw(password.encode(), bcrypt.gensalt()).decode()
+
+    @staticmethod
+    def verify_password(password: str, password_hash: str) -> bool:
+        """Verify a password against its hash.
+
+        Args:
+            password: Plain text password
+            password_hash: Stored hash
+
+        Returns:
+            True if password matches
+        """
+        return bcrypt.checkpw(password.encode(), password_hash.encode())
+
+    def create_user(
+        self,
+        email: str,
+        password: str,
+        role: str = "user",
+    ) -> Optional[Dict]:
+        """Create a new user.
+
+        Args:
+            email: User email
+            password: Plain text password
+            role: User role ('admin' or 'user')
+
+        Returns:
+            Created user dict or None if email exists
+        """
+        password_hash = self.hash_password(password)
+
+        try:
+            with self.get_conn() as conn:
+                with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                    cursor.execute(
+                        """
+                        INSERT INTO users (email, password_hash, role)
+                        VALUES (%s, %s, %s)
+                        RETURNING id, email, role, created_at
+                        """,
+                        (email, password_hash, role),
+                    )
+                    user = cursor.fetchone()
+                conn.commit()
+                return dict(user) if user else None
+        except psycopg2.errors.UniqueViolation:
+            return None
+
+    def authenticate_user(self, email: str, password: str) -> Optional[Dict]:
+        """Authenticate a user by email and password.
+
+        Args:
+            email: User email
+            password: Plain text password
+
+        Returns:
+            User dict if authenticated, None otherwise
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    "SELECT * FROM users WHERE email = %s",
+                    (email,),
+                )
+                user = cursor.fetchone()
+
+                if user and self.verify_password(password, user["password_hash"]):
+                    return {
+                        "id": user["id"],
+                        "email": user["email"],
+                        "role": user["role"],
+                        "created_at": user["created_at"],
+                    }
+                return None
+
+    def get_user_by_id(self, user_id: int) -> Optional[Dict]:
+        """Get a user by ID.
+
+        Args:
+            user_id: User ID
+
+        Returns:
+            User dict or None
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    "SELECT id, email, role, created_at FROM users WHERE id = %s",
+                    (user_id,),
+                )
+                user = cursor.fetchone()
+                return dict(user) if user else None
+
+    def get_user_by_email(self, email: str) -> Optional[Dict]:
+        """Get a user by email.
+
+        Args:
+            email: User email
+
+        Returns:
+            User dict or None
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    "SELECT id, email, role, created_at FROM users WHERE email = %s",
+                    (email,),
+                )
+                user = cursor.fetchone()
+                return dict(user) if user else None
+
+    def get_all_users(self, limit: int = 100, offset: int = 0) -> List[Dict]:
+        """Get all users (admin only).
+
+        Args:
+            limit: Maximum number of users
+            offset: Offset for pagination
+
+        Returns:
+            List of user dicts
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    """
+                    SELECT id, email, role, created_at
+                    FROM users
+                    ORDER BY created_at DESC
+                    LIMIT %s OFFSET %s
+                    """,
+                    (limit, offset),
+                )
+                return [dict(row) for row in cursor.fetchall()]
+
+    def update_user_role(self, user_id: int, role: str) -> Optional[Dict]:
+        """Update a user's role (admin only).
+
+        Args:
+            user_id: User ID
+            role: New role ('admin' or 'user')
+
+        Returns:
+            Updated user dict or None
+        """
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    """
+                    UPDATE users
+                    SET role = %s, updated_at = CURRENT_TIMESTAMP
+                    WHERE id = %s
+                    RETURNING id, email, role, created_at
+                    """,
+                    (role, user_id),
+                )
+                user = cursor.fetchone()
+            conn.commit()
+            return dict(user) if user else None
+
+    def delete_user(self, user_id: int) -> bool:
+        """Delete a user (admin only).
+
+        Args:
+            user_id: User ID
+
+        Returns:
+            True if deleted
+        """
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute("DELETE FROM users WHERE id = %s", (user_id,))
+                deleted = cursor.rowcount > 0
+            conn.commit()
+            return deleted
+
+    def get_user_count(self) -> int:
+        """Get total user count.
+
+        Returns:
+            Number of users
+        """
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute("SELECT COUNT(*) FROM users")
+                return cursor.fetchone()[0]
+
+    # Tracked Companies Methods
+
+    def add_tracked_company(self, company_name: str) -> Optional[Dict]:
+        """Add a company to the tracking list."""
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                try:
+                    cursor.execute(
+                        "INSERT INTO tracked_companies (company_name) VALUES (%s) RETURNING *",
+                        (company_name,),
+                    )
+                    row = cursor.fetchone()
+                    conn.commit()
+                    return dict(row) if row else None
+                except Exception:
+                    conn.rollback()
+                    return None
+
+    def remove_tracked_company(self, company_name: str) -> bool:
+        """Remove a company from the tracking list."""
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    "DELETE FROM tracked_companies WHERE LOWER(company_name) = LOWER(%s)",
+                    (company_name,),
+                )
+                conn.commit()
+                return cursor.rowcount > 0
+
+    def list_tracked_companies(self) -> List[Dict]:
+        """List all tracked companies."""
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute("SELECT * FROM tracked_companies ORDER BY company_name")
+                return [dict(row) for row in cursor.fetchall()]
+
+    def update_tracked_company(
+        self, company_name: str, patent_count: int
+    ) -> None:
+        """Update the last analysis stats for a tracked company."""
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """UPDATE tracked_companies
+                       SET last_patent_count = %s, last_analysis_at = CURRENT_TIMESTAMP
+                       WHERE LOWER(company_name) = LOWER(%s)""",
+                    (patent_count, company_name),
+                )
+                conn.commit()
+
+    def store_alert(
+        self,
+        company_name: str,
+        alert_type: str,
+        message: str,
+        old_value: float | None = None,
+        new_value: float | None = None,
+    ) -> None:
+        """Record an alert for a significant change."""
+        with self.get_conn() as conn:
+            with conn.cursor() as cursor:
+                cursor.execute(
+                    """INSERT INTO alerts (company_name, alert_type, message, old_value, new_value)
+                       VALUES (%s, %s, %s, %s, %s)""",
+                    (company_name, alert_type, message, old_value, new_value),
+                )
+                conn.commit()
+
+    def list_alerts(self, limit: int = 50) -> List[Dict]:
+        """List recent alerts."""
+        with self.get_conn() as conn:
+            with conn.cursor(cursor_factory=RealDictCursor) as cursor:
+                cursor.execute(
+                    "SELECT * FROM alerts ORDER BY created_at DESC LIMIT %s",
+                    (limit,),
+                )
+                return [dict(row) for row in cursor.fetchall()]

SPARC/llm.py

@@ -1,28 +1,52 @@
-"""LLM integration for patent analysis using Anthropic's Claude."""
+"""LLM integration for patent analysis using OpenRouter."""
 
-from anthropic import Anthropic
-from SPARC import config
+import logging
 from typing import Dict
 
+from openai import OpenAI
+
+from SPARC import config
+from SPARC.database import DatabaseClient
+
+logger = logging.getLogger(__name__)
+
 
 class LLMAnalyzer:
     """Handles LLM-based analysis of patent content."""
 
-    def __init__(self, api_key: str | None = None):
+    def __init__(self, api_key: str | None = None, test_mode: bool = False, use_cache: bool | None = None):
         """Initialize the LLM analyzer.
 
         Args:
-          api_key: Anthropic API key. If None, will attempt to load from config.
+          api_key: OpenRouter API key. If None, will attempt to load from config.
+          test_mode: If True, print prompts instead of making API calls
+          use_cache: If True, check database cache before making API calls.
+                    If None, uses config.use_cache (default: True)
         """
-        self.client = Anthropic(api_key=api_key or config.anthropic_api_key)
-        self.model = "claude-3-5-sonnet-20241022"
+        self.test_mode = test_mode
+        self.use_cache = use_cache if use_cache is not None else config.use_cache
+        self.model = config.model
 
-    def analyze_patent_content(self, patent_content: str, company_name: str) -> str:
+        # Always initialize database client for storage and caching
+        self.db_client = DatabaseClient(config.database_url)
+        self.db_client.initialize_schema()
+
+        # Initialize OpenRouter client if API key is available
+        if (api_key or config.openrouter_api_key) and not test_mode:
+            self.client = OpenAI(
+                api_key=api_key or config.openrouter_api_key,
+                base_url="https://openrouter.ai/api/v1"
+            )
+        else:
+            self.client = None
+
+    def analyze_patent_content(self, patent_content: str, company_name: str, model: str | None = None) -> str:
         """Analyze patent content to estimate company innovation and performance.
 
         Args:
           patent_content: Minimized patent text (abstract, claims, summary)
           company_name: Name of the company for context
+          model: Optional model override (e.g. "openai/gpt-4o"). Defaults to config.
 
         Returns:
           Analysis text describing innovation quality and potential impact
@@ -40,16 +64,76 @@ Patent Content:
 
 Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals about the company's technical direction and competitive advantage."""
 
-        message = self.client.messages.create(
-            model=self.model,
-            max_tokens=1024,
-            messages=[{"role": "user", "content": prompt}],
-        )
+        effective_model = model or self.model
 
-        return message.content[0].text
+        if self.test_mode:
+            logger.debug("TEST MODE - Prompt that would be sent to LLM:\n%s", prompt)
+            return "[TEST MODE - No API call made]"
+
+        # Check cache first
+        if self.use_cache:
+            cached = self.db_client.get_cached_response(
+                prompt=prompt,
+                company_name=company_name,
+                analysis_type="single_patent"
+            )
+            if cached:
+                # Log the cache hit
+                self.db_client.store_message(
+                    prompt=prompt,
+                    response=cached["response"],
+                    company_name=company_name,
+                    analysis_type="single_patent",
+                    model=effective_model,
+                    metadata={
+                        "patent_content_length": len(patent_content),
+                        "cache_hit": True,
+                        "original_message_id": cached["id"]
+                    },
+                    is_cached=True
+                )
+                return cached["response"]
+
+        # Call API if no cache hit and client is available
+        if self.client:
+            response = self.client.chat.completions.create(
+                model=effective_model,
+                max_tokens=1024,
+                messages=[{"role": "user", "content": prompt}],
+            )
+            response_text = response.choices[0].message.content
+
+            # Store in database for future cache lookups
+            self.db_client.store_message(
+                prompt=prompt,
+                response=response_text,
+                company_name=company_name,
+                analysis_type="single_patent",
+                model=effective_model,
+                metadata={"patent_content_length": len(patent_content)},
+                token_usage={
+                    "prompt_tokens": response.usage.prompt_tokens,
+                    "completion_tokens": response.usage.completion_tokens,
+                    "total_tokens": response.usage.total_tokens
+                } if hasattr(response, 'usage') else None
+            )
+
+            return response_text
+
+        # No API client available - store prompt for later processing
+        placeholder = "[NO API] Prompt stored in database. Configure OPENROUTER_API_KEY to enable analysis."
+        self.db_client.store_message(
+            prompt=prompt,
+            response=placeholder,
+            company_name=company_name,
+            analysis_type="single_patent",
+            model=effective_model,
+            metadata={"patent_content_length": len(patent_content), "pending": True}
+        )
+        return placeholder
 
     def analyze_patent_portfolio(
-        self, patents_data: list[Dict[str, str]], company_name: str
+        self, patents_data: list[Dict[str, str]], company_name: str, model: str | None = None
     ) -> str:
         """Analyze multiple patents to estimate overall company performance.
 
@@ -84,10 +168,81 @@ Patent Portfolio:
 
 Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the company's innovation strength and performance outlook."""
 
-        message = self.client.messages.create(
-            model=self.model,
-            max_tokens=2048,
-            messages=[{"role": "user", "content": prompt}],
-        )
+        effective_model = model or self.model
 
-        return message.content[0].text
+        if self.test_mode:
+            logger.debug("TEST MODE - Portfolio prompt:\n%s", prompt)
+            return "[TEST MODE]"
+
+        metadata = {
+            "patent_count": len(patents_data),
+            "patent_ids": [p['patent_id'] for p in patents_data],
+            "model": effective_model,
+        }
+
+        # Check cache first
+        if self.use_cache:
+            cached = self.db_client.get_cached_response(
+                prompt=prompt,
+                company_name=company_name,
+                analysis_type="portfolio"
+            )
+            if cached:
+                # Log the cache hit
+                self.db_client.store_message(
+                    prompt=prompt,
+                    response=cached["response"],
+                    company_name=company_name,
+                    analysis_type="portfolio",
+                    model=effective_model,
+                    metadata={
+                        **metadata,
+                        "cache_hit": True,
+                        "original_message_id": cached["id"]
+                    },
+                    is_cached=True
+                )
+                return cached["response"]
+
+        # Call API if no cache hit and client is available
+        if self.client:
+            try:
+                response = self.client.chat.completions.create(
+                    model=effective_model,
+                    max_tokens=2048,
+                    messages=[{"role": "user", "content": prompt}],
+                )
+
+                response_text = response.choices[0].message.content
+
+                # Store in database for future cache lookups
+                self.db_client.store_message(
+                    prompt=prompt,
+                    response=response_text,
+                    company_name=company_name,
+                    analysis_type="portfolio",
+                    model=effective_model,
+                    metadata=metadata,
+                    token_usage={
+                        "prompt_tokens": response.usage.prompt_tokens,
+                        "completion_tokens": response.usage.completion_tokens,
+                        "total_tokens": response.usage.total_tokens
+                    } if hasattr(response, 'usage') else None
+                )
+
+                return response_text
+            except AttributeError:
+                return prompt
+
+        # No API client available - store prompt for later processing
+        placeholder = "[NO API] Prompt stored in database. Configure OPENROUTER_API_KEY to enable analysis."
+        self.db_client.store_message(
+            prompt=prompt,
+            response=placeholder,
+            company_name=company_name,
+            analysis_type="portfolio",
+            model=effective_model,
+            metadata={**metadata, "pending": True}
+        )
+        return placeholder
+       

SPARC/scheduler.py

@@ -0,0 +1,109 @@
+"""Scheduled patent analysis for tracked companies.
+
+Uses APScheduler to periodically re-analyze tracked companies and
+detect significant changes in patent counts.
+"""
+
+import logging
+import os
+
+from SPARC import config
+from SPARC.analyzer import CompanyAnalyzer
+from SPARC.database import DatabaseClient
+
+logger = logging.getLogger(__name__)
+
+# Configurable via environment variable (in hours, default 24)
+SCHEDULE_INTERVAL_HOURS = int(os.getenv("SCHEDULE_INTERVAL_HOURS", "24"))
+
+# Patent count change threshold (percentage) to trigger an alert
+CHANGE_THRESHOLD_PERCENT = int(os.getenv("CHANGE_THRESHOLD_PERCENT", "20"))
+
+
+def run_scheduled_analysis() -> None:
+    """Re-analyze all tracked companies and check for significant changes."""
+    db = DatabaseClient(config.database_url)
+    db.connect()
+    db.initialize_schema()
+
+    tracked = db.list_tracked_companies()
+    if not tracked:
+        logger.info("No tracked companies configured; skipping scheduled analysis")
+        return
+
+    logger.info("Running scheduled analysis for %d tracked companies", len(tracked))
+
+    analyzer = CompanyAnalyzer(db_client=db)
+
+    for company_row in tracked:
+        name = company_row["company_name"]
+        old_count = company_row.get("last_patent_count", 0) or 0
+
+        try:
+            result = analyzer._analyze_company_safe(name)
+
+            if result.success:
+                new_count = result.patent_count
+
+                # Update tracking record
+                db.update_tracked_company(name, new_count)
+
+                # Check for significant change
+                if old_count > 0:
+                    delta_pct = abs(new_count - old_count) / old_count * 100
+                    if delta_pct >= CHANGE_THRESHOLD_PERCENT:
+                        direction = "increased" if new_count > old_count else "decreased"
+                        message = (
+                            f"Patent count for {name} {direction} by {delta_pct:.0f}% "
+                            f"({old_count} -> {new_count})"
+                        )
+                        logger.warning("ALERT: %s", message)
+                        db.store_alert(
+                            company_name=name,
+                            alert_type="patent_count_change",
+                            message=message,
+                            old_value=old_count,
+                            new_value=new_count,
+                        )
+                elif new_count > 0:
+                    # First analysis -- record baseline
+                    logger.info("Baseline for %s: %d patents", name, new_count)
+            else:
+                logger.warning("Scheduled analysis failed for %s: %s", name, result.error)
+
+        except Exception as e:
+            logger.error("Error analyzing tracked company %s: %s", name, e)
+
+    db.close()
+    logger.info("Scheduled analysis complete")
+
+
+def start_scheduler() -> None:
+    """Start the APScheduler background scheduler.
+
+    Safe to call at application startup. If apscheduler is not installed,
+    the function logs a warning and returns without starting anything.
+    """
+    try:
+        from apscheduler.schedulers.background import BackgroundScheduler
+    except ImportError:
+        logger.warning(
+            "apscheduler not installed; scheduled analysis disabled. "
+            "Install with: pip install apscheduler"
+        )
+        return
+
+    scheduler = BackgroundScheduler()
+    scheduler.add_job(
+        run_scheduled_analysis,
+        "interval",
+        hours=SCHEDULE_INTERVAL_HOURS,
+        id="scheduled_patent_analysis",
+        replace_existing=True,
+    )
+    scheduler.start()
+    logger.info(
+        "Scheduled patent analysis started (every %d hours, threshold %d%%)",
+        SCHEDULE_INTERVAL_HOURS,
+        CHANGE_THRESHOLD_PERCENT,
+    )

SPARC/serp_api.py

@@ -1,67 +1,134 @@
-import serpapi
-from SPARC import config
+import io
+import logging
 import re
+from datetime import datetime, timedelta
+from typing import Dict
+
 import pdfplumber  # pip install pdfplumber
 import requests
-from typing import Dict
-from SPARC.types import Patents, Patent
+import serpapi
+
+from SPARC import config
+from SPARC.storage import StorageBackend, get_storage_backend
+from SPARC.types import Patent, Patents
+
+logger = logging.getLogger(__name__)
+
+# Module-level storage instance (lazy-initialized)
+_storage: StorageBackend | None = None
+
+
+def _get_storage() -> StorageBackend:
+    global _storage
+    if _storage is None:
+        _storage = get_storage_backend()
+    return _storage
+
 
 class SERP:
-  def query(company: str) -> Patents:
+  def query(company: str, days_back: int = None) -> Patents:
+    """Query Google Patents for a company's recent patents.
+
+    Args:
+        company: Name of the company to search for
+        days_back: Number of days to look back for patents (default from config)
+
+    Returns:
+        Patents object containing list of patents with PDF links
+
+    Note:
+        Patents without PDF download links are skipped. This occurs when
+        Google Patents doesn't have a PDF available for a particular patent
+        (e.g., recently filed patents, certain international patents, or
+        patents with restricted access). The returned count may be lower
+        than the requested number of results.
+    """
+    if days_back is None:
+        days_back = config.patent_search_days
+    end_date = datetime.now()
+    start_date = end_date - timedelta(days=days_back)
+    date_filter = f"cdr:1,cd_min:{start_date.strftime('%-m/%-d/%Y')},cd_max:{end_date.strftime('%-m/%-d/%Y')}"
+
     # Make API call
     params = {
       "engine": "google_patents",
       "q": company,
       "num": 10,
       "filter": 1,
-      "tbs": "cdr:1,cd_min:10/28/2025,cd_max:11/4/2025",
+      "tbs": date_filter,
       "api_key": config.api_key,
     }
+    logger.info("Querying Google Patents for '%s' (last %d days)", company, days_back)
     search = serpapi.search(params)
-    # Convert data into a list of publicationID
+    # Convert results to Patent objects, skipping any without PDF links
     patent_ids = []
     list_of_patents = search["organic_results"]
     for patent in list_of_patents:
-        patent_ids.append(Patent(patent_id=patent["publication_number"], pdf_link=patent["pdf"], summary=None))
+        pdf_link = patent.get("pdf")
+        if pdf_link:
+            patent_ids.append(Patent(patent_id=patent["publication_number"], pdf_link=pdf_link, summary=None))
+        else:
+            logger.debug("Skipping patent %s (no PDF link)", patent.get("publication_number", "unknown"))
 
+    logger.info("Found %d patents with PDF links for '%s'", len(patent_ids), company)
     return Patents(patents=patent_ids)
 
   def save_patents(patent: Patent) -> Patent:
-    """
-    Save the patent PDF to the patents folder
-    
+    """Save the patent PDF to storage, skipping download if already cached.
+
+    Uses the configured storage backend (local filesystem or S3).
+
     Args:
       patent: Patent object
 
     Returns:
       Patent object with updated PDF path
     """
-    response = requests.get(patent.pdf_link)
-    print(patent.pdf_link)
-    with open(f"patents/{patent.patent_id}.pdf", "wb") as f:
-      f.write(response.content)
-    
-    patent.pdf_path = f"patents/{patent.patent_id}.pdf"
+    storage = _get_storage()
+    key = f"{patent.patent_id}.pdf"
+
+    if not storage.exists(key):
+      logger.info("Downloading PDF for %s", patent.patent_id)
+      response = requests.get(patent.pdf_link)
+      storage.write(key, response.content)
+      logger.debug("Saved %d bytes for %s", len(response.content), patent.patent_id)
+    else:
+      logger.debug("Using cached PDF for %s", patent.patent_id)
+
+    patent.pdf_path = storage.path_for(key)
     return patent
 
   def parse_patent_pdf(pdf_path: str) -> Dict:
     """Extract structured sections from patent PDF.
 
     Extracts all major sections from a patent PDF including abstract,
-    claims, summary, and detailed description.
+    claims, summary, and detailed description. Supports both local file
+    paths and S3 URIs (s3://bucket/key).
 
     Args:
-      pdf_path: Path to the patent PDF file
+      pdf_path: Local path or S3 URI to the patent PDF file
 
     Returns:
       Dictionary containing all extracted sections
     """
+    logger.debug("Parsing patent PDF: %s", pdf_path)
 
-    with pdfplumber.open(pdf_path) as pdf:
+    if pdf_path.startswith("s3://"):
+      # Read from S3 via storage backend
+      storage = _get_storage()
+      # Extract key from "s3://bucket/key"
+      key = pdf_path.split("/", 3)[-1]
+      data = storage.read(key)
+      pdf_file: io.BytesIO | str = io.BytesIO(data)
+    else:
+      pdf_file = pdf_path
+
+    with pdfplumber.open(pdf_file) as pdf:
       # Extract all text
       full_text = ""
       for page in pdf.pages:
         full_text += page.extract_text() + "\n"
+      logger.debug("Extracted text from %d pages (%d chars)", len(pdf.pages), len(full_text))
 
     # Define section patterns (common in patents)
     sections = {

SPARC/storage.py

@@ -0,0 +1,171 @@
+"""Patent PDF storage abstraction.
+
+Provides a unified interface for reading and writing patent PDF files,
+with pluggable backends for local filesystem and S3-compatible object
+storage (e.g., MinIO, AWS S3).
+"""
+
+import logging
+import os
+from abc import ABC, abstractmethod
+
+from SPARC import config
+
+logger = logging.getLogger(__name__)
+
+
+class StorageBackend(ABC):
+    """Abstract base class for patent PDF storage."""
+
+    @abstractmethod
+    def read(self, key: str) -> bytes:
+        """Read a file by key.
+
+        Args:
+            key: Storage key (e.g., "US-12345678-B2.pdf")
+
+        Returns:
+            File contents as bytes.
+
+        Raises:
+            FileNotFoundError: If the file does not exist.
+        """
+
+    @abstractmethod
+    def write(self, key: str, data: bytes) -> None:
+        """Write data to storage.
+
+        Args:
+            key: Storage key (e.g., "US-12345678-B2.pdf")
+            data: File contents as bytes.
+        """
+
+    @abstractmethod
+    def exists(self, key: str) -> bool:
+        """Check if a file exists in storage.
+
+        Args:
+            key: Storage key.
+
+        Returns:
+            True if the file exists and has non-zero size.
+        """
+
+    @abstractmethod
+    def path_for(self, key: str) -> str:
+        """Return a path or URI suitable for downstream consumers.
+
+        For local storage this is a filesystem path; for S3 it is the
+        object key (callers that need a local file should use read()
+        and write to a temporary location).
+        """
+
+
+class LocalStorageBackend(StorageBackend):
+    """Store patent PDFs on the local filesystem under a directory."""
+
+    def __init__(self, base_dir: str = "patents"):
+        self.base_dir = base_dir
+        os.makedirs(self.base_dir, exist_ok=True)
+
+    def _full_path(self, key: str) -> str:
+        return os.path.join(self.base_dir, key)
+
+    def read(self, key: str) -> bytes:
+        path = self._full_path(key)
+        if not os.path.exists(path):
+            raise FileNotFoundError(f"File not found: {path}")
+        with open(path, "rb") as f:
+            return f.read()
+
+    def write(self, key: str, data: bytes) -> None:
+        path = self._full_path(key)
+        os.makedirs(os.path.dirname(path) or self.base_dir, exist_ok=True)
+        with open(path, "wb") as f:
+            f.write(data)
+        logger.debug("Wrote %d bytes to %s", len(data), path)
+
+    def exists(self, key: str) -> bool:
+        path = self._full_path(key)
+        return os.path.exists(path) and os.path.getsize(path) > 0
+
+    def path_for(self, key: str) -> str:
+        return self._full_path(key)
+
+
+class S3StorageBackend(StorageBackend):
+    """Store patent PDFs in an S3-compatible bucket."""
+
+    def __init__(
+        self,
+        bucket: str,
+        endpoint_url: str = "",
+        access_key: str = "",
+        secret_key: str = "",
+    ):
+        import boto3
+
+        kwargs: dict = {}
+        if endpoint_url:
+            kwargs["endpoint_url"] = endpoint_url
+        if access_key and secret_key:
+            kwargs["aws_access_key_id"] = access_key
+            kwargs["aws_secret_access_key"] = secret_key
+
+        self.s3 = boto3.client("s3", **kwargs)
+        self.bucket = bucket
+
+        # Ensure bucket exists (useful for MinIO local dev)
+        try:
+            self.s3.head_bucket(Bucket=self.bucket)
+        except Exception:
+            try:
+                self.s3.create_bucket(Bucket=self.bucket)
+                logger.info("Created S3 bucket: %s", self.bucket)
+            except Exception as e:
+                logger.warning("Could not create bucket %s: %s", self.bucket, e)
+
+    def read(self, key: str) -> bytes:
+        try:
+            response = self.s3.get_object(Bucket=self.bucket, Key=key)
+            return response["Body"].read()
+        except self.s3.exceptions.NoSuchKey:
+            raise FileNotFoundError(f"S3 object not found: s3://{self.bucket}/{key}")
+        except Exception as e:
+            if "NoSuchKey" in str(e) or "404" in str(e):
+                raise FileNotFoundError(f"S3 object not found: s3://{self.bucket}/{key}")
+            raise
+
+    def write(self, key: str, data: bytes) -> None:
+        self.s3.put_object(
+            Bucket=self.bucket,
+            Key=key,
+            Body=data,
+            ContentType="application/pdf",
+        )
+        logger.debug("Wrote %d bytes to s3://%s/%s", len(data), self.bucket, key)
+
+    def exists(self, key: str) -> bool:
+        try:
+            response = self.s3.head_object(Bucket=self.bucket, Key=key)
+            return response["ContentLength"] > 0
+        except Exception:
+            return False
+
+    def path_for(self, key: str) -> str:
+        return f"s3://{self.bucket}/{key}"
+
+
+def get_storage_backend() -> StorageBackend:
+    """Factory: return the configured storage backend instance."""
+    backend = config.storage_backend.lower()
+    if backend == "s3":
+        logger.info("Using S3 storage backend (bucket=%s)", config.s3_bucket)
+        return S3StorageBackend(
+            bucket=config.s3_bucket,
+            endpoint_url=config.s3_endpoint_url,
+            access_key=config.s3_access_key,
+            secret_key=config.s3_secret_key,
+        )
+    logger.info("Using local storage backend")
+    return LocalStorageBackend()

SPARC/types.py

@@ -1,9 +1,10 @@
-from dataclasses import dataclass
+from dataclasses import dataclass, field
+from datetime import datetime
 
 
 @dataclass
 class Patent:
-    patent_id: int
+    patent_id: str
     pdf_link: str
     pdf_path: str | None = None
     summary: dict | None = None
@@ -12,3 +13,27 @@ class Patent:
 @dataclass
 class Patents:
     patents: list[Patent]
+
+
+@dataclass
+class CompanyAnalysisResult:
+    """Result of analyzing a single company's patent portfolio."""
+
+    company_name: str
+    analysis: str
+    patent_count: int
+    success: bool
+    error: str | None = None
+    model: str | None = None
+    timestamp: datetime = field(default_factory=datetime.now)
+
+
+@dataclass
+class BatchAnalysisResult:
+    """Result of batch analyzing multiple companies."""
+
+    results: list[CompanyAnalysisResult]
+    total_companies: int
+    successful: int
+    failed: int
+    timestamp: datetime = field(default_factory=datetime.now)

SPARC/webhooks.py

@@ -0,0 +1,139 @@
+"""Webhook notifications for job completion and alert events.
+
+Sends JSON payloads to configured webhook URLs with retry logic.
+Supports generic HTTP POST and Slack-compatible text payloads.
+"""
+
+import logging
+import os
+import time
+from datetime import datetime
+from typing import Any
+
+import requests
+
+logger = logging.getLogger(__name__)
+
+# Comma-separated list of webhook URLs (env var based config)
+_WEBHOOK_URLS_RAW = os.getenv("WEBHOOK_URLS", "")
+WEBHOOK_URLS: list[str] = [
+    url.strip() for url in _WEBHOOK_URLS_RAW.split(",") if url.strip()
+]
+
+MAX_RETRIES = 3
+BACKOFF_BASE = 2  # seconds
+
+
+def _is_slack_url(url: str) -> bool:
+    """Check if a URL looks like a Slack incoming webhook."""
+    return "hooks.slack.com" in url or "discord.com/api/webhooks" in url
+
+
+def _build_payload(event_type: str, data: dict[str, Any], slack: bool = False) -> dict:
+    """Build the webhook payload.
+
+    Args:
+        event_type: Type of event (e.g., "job_completed", "alert")
+        data: Event-specific data
+        slack: If True, wrap in Slack-compatible ``text`` format
+
+    Returns:
+        JSON-serializable payload dict
+    """
+    payload = {
+        "event": event_type,
+        "timestamp": datetime.utcnow().isoformat() + "Z",
+        **data,
+    }
+
+    if slack:
+        # Build a human-readable summary for Slack/Discord
+        lines = [f"*[SPARC] {event_type}*"]
+        for key, value in data.items():
+            lines.append(f"  {key}: {value}")
+        return {"text": "\n".join(lines)}
+
+    return payload
+
+
+def _send_with_retry(url: str, payload: dict) -> bool:
+    """Send a POST request with exponential backoff retry.
+
+    Args:
+        url: Webhook URL
+        payload: JSON payload to send
+
+    Returns:
+        True if delivered successfully, False after all retries exhausted
+    """
+    for attempt in range(1, MAX_RETRIES + 1):
+        try:
+            response = requests.post(url, json=payload, timeout=10)
+            if response.status_code < 300:
+                logger.debug("Webhook delivered to %s (attempt %d)", url, attempt)
+                return True
+            logger.warning(
+                "Webhook %s returned %d (attempt %d/%d)",
+                url, response.status_code, attempt, MAX_RETRIES,
+            )
+        except requests.RequestException as e:
+            logger.warning(
+                "Webhook delivery failed for %s (attempt %d/%d): %s",
+                url, attempt, MAX_RETRIES, e,
+            )
+
+        if attempt < MAX_RETRIES:
+            wait = BACKOFF_BASE ** attempt
+            time.sleep(wait)
+
+    logger.error("Webhook permanently failed for %s after %d attempts", url, MAX_RETRIES)
+    return False
+
+
+def notify(event_type: str, data: dict[str, Any]) -> None:
+    """Fire all configured webhooks for an event.
+
+    Safe to call even when no webhooks are configured (returns immediately).
+
+    Args:
+        event_type: Event identifier (e.g., "job_completed", "patent_alert")
+        data: Event data to include in the payload
+    """
+    if not WEBHOOK_URLS:
+        return
+
+    for url in WEBHOOK_URLS:
+        slack = _is_slack_url(url)
+        payload = _build_payload(event_type, data, slack=slack)
+        _send_with_retry(url, payload)
+
+
+def notify_job_completed(
+    job_id: str,
+    status: str,
+    total_companies: int,
+    successful: int,
+    failed: int,
+) -> None:
+    """Send notification when a batch job completes."""
+    notify("job_completed", {
+        "job_id": job_id,
+        "status": status,
+        "total_companies": total_companies,
+        "successful": successful,
+        "failed": failed,
+        "summary": f"Batch job {job_id}: {successful}/{total_companies} succeeded",
+    })
+
+
+def notify_alert(
+    company_name: str,
+    alert_type: str,
+    message: str,
+) -> None:
+    """Send notification for a tracked company alert."""
+    notify("patent_alert", {
+        "company_name": company_name,
+        "alert_type": alert_type,
+        "message": message,
+    })

docker-compose.yml

@@ -0,0 +1,93 @@
+services:
+  postgres:
+    image: postgres:16-alpine
+    container_name: sparc-postgres
+    environment:
+      POSTGRES_USER: ${POSTGRES_USER}
+      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
+      POSTGRES_DB: ${POSTGRES_DB}
+    ports:
+      - "5432:5432"
+    volumes:
+      - postgres_data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER}"]
+      interval: 5s
+      timeout: 5s
+      retries: 5
+    restart: unless-stopped
+
+  init-db:
+    image: gitea.leeworks.dev/0xwheatyz/sparc:latest
+    build: .
+    container_name: sparc-init-db
+    command: python scripts/init_database.py
+    environment:
+      DATABASE_URL: postgresql://${POSTGRES_USER}:${POSTGRES_PASSWORD}@postgres:5432/${POSTGRES_DB}
+    depends_on:
+      postgres:
+        condition: service_healthy
+    restart: "no"
+
+  api:
+    image: gitea.leeworks.dev/0xwheatyz/sparc:latest
+    build: .
+    container_name: sparc-api
+    command: uvicorn SPARC.api:app --host 0.0.0.0 --port 8000
+    environment:
+      API_KEY: ${API_KEY}
+      OPENROUTER_API_KEY: ${OPENROUTER_API_KEY}
+      DATABASE_URL: postgresql://${POSTGRES_USER}:${POSTGRES_PASSWORD}@postgres:5432/${POSTGRES_DB}
+      USE_CACHE: "true"
+      JWT_SECRET: ${JWT_SECRET:-sparc-secret-key-change-in-production}
+      CORS_ORIGINS: ${CORS_ORIGINS:-}
+      APP_ENV: ${APP_ENV:-development}
+      ROOT_PATH: ""
+    ports:
+      - "8000:8000"
+    depends_on:
+      postgres:
+        condition: service_healthy
+      init-db:
+        condition: service_completed_successfully
+    volumes:
+      - patent_data:/app/patents
+    restart: unless-stopped
+
+  # Optional: MinIO for S3-compatible local object storage
+  # Enable by setting STORAGE_BACKEND=s3 in .env
+  minio:
+    image: minio/minio:latest
+    container_name: sparc-minio
+    command: server /data --console-address ":9001"
+    environment:
+      MINIO_ROOT_USER: ${AWS_ACCESS_KEY_ID:-minioadmin}
+      MINIO_ROOT_PASSWORD: ${AWS_SECRET_ACCESS_KEY:-minioadmin}
+    ports:
+      - "9000:9000"
+      - "9001:9001"
+    volumes:
+      - minio_data:/data
+    healthcheck:
+      test: ["CMD", "mc", "ready", "local"]
+      interval: 10s
+      timeout: 5s
+      retries: 3
+    restart: unless-stopped
+    profiles:
+      - s3
+
+  dashboard:
+    image: gitea.leeworks.dev/0xwheatyz/sparc:frontend-latest
+    build: ./frontend
+    container_name: sparc-dashboard
+    ports:
+      - "8080:80"
+    depends_on:
+      - api
+    restart: unless-stopped
+
+volumes:
+  postgres_data:
+  patent_data:
+  minio_data:

docs/CONTAINER_REGISTRY.md

@@ -0,0 +1,188 @@
+# Container Registry and CI/CD Setup
+
+This document explains how to build and push Docker images using Gitea Actions and the Gitea Container Registry.
+
+## Overview
+
+The SPARC project uses Gitea Actions (GitHub Actions-compatible) to automatically build and push Docker images to the Gitea Container Registry whenever code is pushed to the repository.
+
+## Workflow Configuration
+
+The workflow is defined in `.gitea/workflows/build.yaml` and automatically:
+- Builds the Docker image from the `Dockerfile`
+- Tags the image appropriately based on the git ref (branch/tag)
+- Pushes to the Gitea Container Registry at `10.0.1.10`
+
+### Triggers
+
+The workflow runs on:
+- **Push to main branch**: Builds and tags with commit SHA + `latest`
+- **Push of tags**: Builds and tags with the tag name + `latest`
+- **Manual dispatch**: Can be triggered manually from Gitea UI
+
+### Image Naming
+
+Images are pushed to: `10.0.1.10/0xwheatyz/sparc:<tag>`
+
+- Main branch commits: `10.0.1.10/0xwheatyz/sparc:<sha>` and `10.0.1.10/0xwheatyz/sparc:latest`
+- Tags: `10.0.1.10/0xwheatyz/sparc:<tag-name>` and `10.0.1.10/0xwheatyz/sparc:latest`
+- Other branches: `10.0.1.10/0xwheatyz/sparc:<branch-name>`
+
+## Prerequisites
+
+### 1. Enable Container Registry in Gitea
+
+The Gitea instance must have the Container Registry (Packages) feature enabled:
+
+1. Access Gitea as administrator
+2. Go to Site Administration > Configuration
+3. Find "Packages" section
+4. Ensure packages/container registry is enabled
+
+### 2. Create Personal Access Token
+
+The workflow needs a personal access token with package write permissions:
+
+1. In Gitea UI, click your profile → Settings
+2. Go to Applications → Manage Access Tokens
+3. Click "Generate New Token"
+4. Give it a descriptive name (e.g., "Actions Container Registry")
+5. Select scopes:
+   - `write:package` (required)
+   - `read:package` (required)
+6. Click "Generate Token"
+7. **Copy the token immediately** (you won't see it again)
+
+### 3. Add Token as Repository Secret
+
+1. Go to your repository in Gitea
+2. Click Settings → Secrets
+3. Click "Add Secret"
+4. Name: `GITEA_TOKEN`
+5. Value: Paste the personal access token
+6. Click "Add Secret"
+
+## Usage
+
+### Automatic Builds
+
+Once configured, the workflow runs automatically:
+
+```bash
+# Push to main branch - triggers build
+git add .
+git commit -m "feat: add new feature"
+git push origin main
+
+# Create and push a tag - triggers build with tag
+git tag v1.0.0
+git push origin v1.0.0
+```
+
+### Manual Builds
+
+You can also trigger builds manually:
+
+1. Go to repository → Actions
+2. Click on "Build and Push Docker Image" workflow
+3. Click "Run workflow"
+4. Select the branch
+5. Click "Run workflow"
+
+### Monitor Build Progress
+
+1. Go to repository → Actions
+2. Click on the running workflow
+3. View logs for each step
+
+## Pulling Images
+
+Once built, images can be pulled from the registry:
+
+```bash
+# Log in to registry
+docker login 10.0.1.10 -u your-username
+
+# Pull the latest image
+docker pull 10.0.1.10/0xwheatyz/sparc:latest
+
+# Pull a specific tag
+docker pull 10.0.1.10/0xwheatyz/sparc:v1.0.0
+
+# Pull a specific commit
+docker pull 10.0.1.10/0xwheatyz/sparc:abc1234
+```
+
+## Troubleshooting
+
+### Workflow Fails at Login Step
+
+**Error**: `Error response from daemon: login attempt to http://10.0.1.10/v2/ failed with status: 404 Not Found`
+
+**Solution**: Container registry is not enabled in Gitea. Contact administrator to enable packages feature.
+
+### Workflow Fails with 401 Unauthorized
+
+**Error**: `unauthorized: authentication required`
+
+**Solutions**:
+1. Verify `GITEA_TOKEN` secret exists and is correct
+2. Verify token has `write:package` and `read:package` scopes
+3. Regenerate token if it has expired
+
+### Workflow Fails at Push Step
+
+**Error**: `denied: permission denied`
+
+**Solutions**:
+1. Ensure your user account has write access to the repository
+2. Verify the token has the correct permissions
+3. Check if the repository owner matches the registry path
+
+### Image Not Appearing in Packages
+
+**Check**:
+1. Go to repository → Packages tab
+2. If no packages appear, check workflow logs for errors
+3. Verify the image was successfully pushed (check workflow output)
+
+## Advanced Configuration
+
+### Using a Different Registry
+
+To push to a different container registry (e.g., Docker Hub, GHCR):
+
+1. Update the `REGISTRY` variable in `.gitea/workflows/build.yaml`
+2. Update the login step with appropriate credentials
+3. Add registry credentials as secrets
+
+### Building Multi-platform Images
+
+To build for multiple architectures:
+
+```yaml
+- name: Build Docker image
+  run: |
+    docker buildx build \
+      --platform linux/amd64,linux/arm64 \
+      -t ${{ steps.tags.outputs.IMAGE_TAG }} \
+      --push .
+```
+
+### Adding Build Arguments
+
+To pass build arguments:
+
+```yaml
+- name: Build Docker image
+  run: |
+    docker build \
+      --build-arg VERSION=${{ gitea.sha_short }} \
+      -t ${{ steps.tags.outputs.IMAGE_TAG }} .
+```
+
+## References
+
+- [Gitea Actions Documentation](https://docs.gitea.com/usage/actions/overview)
+- [Gitea Packages Documentation](https://docs.gitea.com/usage/packages/overview)
+- [GitHub Actions Syntax](https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions) (Gitea Actions compatible)

docs/DATABASE_MODE.md

@@ -0,0 +1,331 @@
+# Database Storage and Caching
+
+This document explains how SPARC uses PostgreSQL for storing LLM messages, enabling response caching and analytics.
+
+## Overview
+
+SPARC stores all LLM interactions in PostgreSQL, providing:
+
+- **Response Caching**: Avoid redundant API calls for previously analyzed patents
+- **Analytics**: Track usage patterns, token consumption, and analysis history
+- **Persistence**: Maintain analysis history across sessions
+
+SPARC supports two cache modes:
+
+1. **Cache Mode** (default, `USE_CACHE=true`): Check database for cached responses before making API calls
+2. **Fresh Mode** (`USE_CACHE=false`): Always make fresh API calls (still stores results in database)
+
+## Setup
+
+### 1. Start the Database
+
+Use docker-compose to start the PostgreSQL database:
+
+```bash
+docker-compose up -d postgres
+```
+
+This will start a PostgreSQL instance accessible at `localhost:5432`.
+
+### 2. Initialize the Database Schema
+
+Run the initialization script to create the necessary tables:
+
+```bash
+python scripts/init_database.py
+```
+
+This creates the `llm_messages` table and indexes for efficient querying.
+
+### 3. Configure Environment Variables
+
+Create a `.env` file (or copy from `.env.example`):
+
+```bash
+cp .env.example .env
+```
+
+Edit `.env` and set:
+
+```env
+# Database connection (required)
+DATABASE_URL=postgresql://postgres:postgres@localhost:5432/sparc
+
+# Cache mode: use cached responses when available
+USE_CACHE=true
+
+# API key for fresh LLM calls
+OPENROUTER_API_KEY=your_openrouter_key_here
+```
+
+## Usage
+
+### Running with Cache Mode (Default)
+
+Set `USE_CACHE=true` in your `.env` file, then run the application normally:
+
+```bash
+python main.py
+```
+
+The application will:
+- Check the database for cached responses matching the request
+- If found, return the cached response (no API call)
+- If not found, make an API call and store the response for future use
+
+### Running with Fresh Mode
+
+Set `USE_CACHE=false` in your `.env` file to always get fresh responses:
+
+```bash
+python main.py
+```
+
+The application will:
+- Always send messages to OpenRouter for real LLM responses
+- Store all responses in the database
+- Useful when you need the latest analysis or want to refresh cached data
+
+## Viewing Analytics
+
+### View Message Statistics
+
+```bash
+python scripts/view_analytics.py
+```
+
+Options:
+- `--days N`: Analyze messages from the last N days (default: 30)
+
+Example output:
+```
+SPARC Analytics - Last 30 days
+======================================================================
+
+Total Messages: 45
+
+Messages by Company:
+  nvidia: 25
+  intel: 12
+  amd: 8
+
+Messages by Analysis Type:
+  portfolio: 30
+  single_patent: 15
+
+======================================================================
+```
+
+### View Stored Messages
+
+```bash
+python scripts/view_messages.py
+```
+
+Options:
+- `--company COMPANY`: Filter by company name
+- `--type TYPE`: Filter by analysis type (single_patent or portfolio)
+- `--limit N`: Maximum number of messages to display (default: 10)
+
+Examples:
+```bash
+# View last 10 messages
+python scripts/view_messages.py
+
+# View all messages for nvidia
+python scripts/view_messages.py --company nvidia --limit 100
+
+# View portfolio analyses only
+python scripts/view_messages.py --type portfolio
+```
+
+## Database Schema
+
+### llm_messages Table
+
+| Column | Type | Description |
+|--------|------|-------------|
+| id | SERIAL | Primary key |
+| timestamp | TIMESTAMP | When the message was created |
+| company_name | VARCHAR(255) | Company being analyzed |
+| analysis_type | VARCHAR(50) | Type of analysis (single_patent, portfolio) |
+| model | VARCHAR(100) | LLM model identifier |
+| prompt | TEXT | The full prompt sent to the LLM |
+| response | TEXT | The response from the LLM |
+| metadata | JSONB | Additional metadata (patent IDs, content length, etc.) |
+| token_usage | JSONB | Token usage statistics (when available) |
+| created_at | TIMESTAMP | Record creation timestamp |
+
+### Indexes
+
+- `idx_messages_timestamp`: Speeds up time-based queries
+- `idx_messages_company`: Speeds up company-specific queries
+
+## Docker Compose
+
+The included `docker-compose.yml` provides:
+
+1. **PostgreSQL Database**:
+   - Image: `postgres:16-alpine`
+   - Port: `5432`
+   - Credentials: postgres/postgres
+   - Database: sparc
+   - Persistent storage via volume
+
+2. **Application Container** (optional):
+   - Builds from Dockerfile
+   - Connects to PostgreSQL
+   - Mounts current directory
+
+### Start Services
+
+```bash
+# Start just the database
+docker-compose up -d postgres
+
+# Start everything
+docker-compose up -d
+
+# View logs
+docker-compose logs -f
+
+# Stop services
+docker-compose down
+
+# Stop and remove volumes (WARNING: deletes data)
+docker-compose down -v
+```
+
+## Toggling Between Modes
+
+You can easily switch between modes by changing the `USE_CACHE` environment variable:
+
+### Quick Toggle (temporary)
+
+```bash
+# Run with caching enabled
+USE_CACHE=true python main.py
+
+# Run with fresh API calls
+USE_CACHE=false python main.py
+```
+
+### Persistent Toggle
+
+Edit your `.env` file:
+
+```env
+# Use cached responses when available (recommended for most use)
+USE_CACHE=true
+
+# Always make fresh API calls
+USE_CACHE=false
+```
+
+## Use Cases
+
+### Cost Optimization with Caching
+
+Cache mode reduces API costs by reusing previous analysis results:
+
+```bash
+USE_CACHE=true python main.py
+```
+
+If the same company/patent combination was analyzed before, the cached response is returned instantly.
+
+### Fresh Analysis
+
+When you need the latest LLM analysis (e.g., after model updates):
+
+```bash
+USE_CACHE=false python main.py
+```
+
+### Collecting Usage Analytics
+
+The database stores all interactions, enabling analytics on:
+- Which companies are analyzed most frequently
+- Types of analyses performed
+- Token usage and costs over time
+- Response caching hit rates
+
+### Development and Debugging
+
+Database storage is useful for:
+- Reviewing actual prompts sent to the LLM
+- Analyzing response patterns
+- Debugging the full pipeline end-to-end
+- Understanding token usage patterns
+
+## Troubleshooting
+
+### Connection Refused
+
+If you get "connection refused" errors:
+
+1. Ensure PostgreSQL is running: `docker-compose ps`
+2. Check the DATABASE_URL in your `.env` file
+3. Wait for the database to be healthy: `docker-compose logs postgres`
+
+### Schema Not Found
+
+If you get "relation does not exist" errors:
+
+1. Run the initialization script: `python scripts/init_database.py`
+2. Verify tables were created: `docker-compose exec postgres psql -U postgres -d sparc -c "\dt"`
+
+### Permission Denied
+
+If you get permission errors:
+
+1. Check your DATABASE_URL credentials match docker-compose.yml
+2. Ensure the database container is running: `docker-compose up -d postgres`
+
+## Advanced Usage
+
+### Direct Database Access
+
+You can access the database directly using psql:
+
+```bash
+docker-compose exec postgres psql -U postgres -d sparc
+```
+
+Example queries:
+
+```sql
+-- View all messages
+SELECT id, company_name, analysis_type, timestamp FROM llm_messages ORDER BY timestamp DESC LIMIT 10;
+
+-- Count messages by company
+SELECT company_name, COUNT(*) FROM llm_messages GROUP BY company_name;
+
+-- View recent prompts
+SELECT prompt FROM llm_messages ORDER BY timestamp DESC LIMIT 5;
+```
+
+### Programmatic Access
+
+You can use the `DatabaseClient` directly in your code:
+
+```python
+from SPARC.database import DatabaseClient
+from SPARC import config
+
+db = DatabaseClient(config.database_url)
+
+# Get messages
+messages = db.get_messages(company_name="nvidia", limit=10)
+
+# Get analytics
+analytics = db.get_analytics(days=7)
+
+# Store a custom message
+db.store_message(
+    prompt="test prompt",
+    response="test response",
+    company_name="test",
+    analysis_type="custom"
+)
+```

docs/DEPLOYMENT.md

@@ -0,0 +1,471 @@
+# SPARC Complete Deployment Guide
+
+This guide provides step-by-step instructions for deploying the SPARC (Semiconductor Patent & Analytics Report Core) application with all features enabled, including SERP API patent retrieval, LLM analysis, database storage, and the web UI.
+
+## Table of Contents
+
+- [Prerequisites](#prerequisites)
+- [Step 1: Clone and Configure](#step-1-clone-and-configure)
+- [Step 2: Start Services with Docker Compose](#step-2-start-services-with-docker-compose)
+- [Step 3: Initialize the Database](#step-3-initialize-the-database)
+- [Step 4: Run the Services](#step-4-run-the-services)
+- [Step 5: Verify Deployment](#step-5-verify-deployment)
+- [Step 6: Using the Application](#step-6-using-the-application)
+- [Step 7: View Stored Data](#step-7-view-stored-data)
+- [Architecture Overview](#architecture-overview)
+- [Environment Variables Reference](#environment-variables-reference)
+- [Production Docker Compose](#production-docker-compose)
+- [Troubleshooting](#troubleshooting)
+
+---
+
+## Prerequisites
+
+1. **Docker & Docker Compose** installed
+2. **API Keys** (you'll need to obtain these):
+   - **SerpAPI Key**: Sign up at https://serpapi.com/ (free tier: 100 searches/month)
+   - **OpenRouter API Key**: Sign up at https://openrouter.ai/ (pay-as-you-go)
+
+---
+
+## Step 1: Clone and Configure
+
+```bash
+git clone <repository-url>
+cd SPARC
+
+# Create environment file
+cp .env.example .env
+```
+
+Edit `.env` with your API keys:
+
+```env
+# Required API Keys
+API_KEY=your_serpapi_key_here
+OPENROUTER_API_KEY=your_openrouter_key_here
+
+# Database Configuration (matches docker-compose.yml)
+DATABASE_URL=postgresql://postgres:postgres@localhost:5432/sparc
+USE_DATABASE=true
+```
+
+---
+
+## Step 2: Start Services with Docker Compose
+
+```bash
+# Start all services (PostgreSQL, API, and Dashboard)
+docker-compose up -d
+
+# Check status
+docker-compose ps
+
+# You should see:
+# - sparc-postgres (healthy)
+# - sparc-api (running on port 8000)
+# - sparc-dashboard (running on port 8080)
+```
+
+The database is automatically initialized by the `init-db` service.
+
+---
+
+## Step 3: Database Schema
+
+The `init-db` service automatically creates the `llm_messages` table with the following schema:
+
+| Column | Type | Purpose |
+|--------|------|---------|
+| `id` | SERIAL | Primary key |
+| `timestamp` | TIMESTAMP | Message creation time |
+| `company_name` | VARCHAR(255) | Company being analyzed |
+| `analysis_type` | VARCHAR(50) | 'single_patent' or 'portfolio' |
+| `model` | VARCHAR(100) | LLM model identifier |
+| `prompt` | TEXT | Full prompt sent to LLM |
+| `response` | TEXT | LLM response |
+| `metadata` | JSONB | Patent IDs, content lengths |
+| `token_usage` | JSONB | prompt/completion/total tokens |
+| `created_at` | TIMESTAMP | Record timestamp |
+
+---
+
+## Step 4: Run the Services
+
+### Option A: Run with Docker Compose (Recommended)
+
+All services are started automatically with `docker-compose up -d` from Step 2.
+
+```bash
+# View logs
+docker-compose logs -f
+
+# View specific service logs
+docker-compose logs -f api
+docker-compose logs -f dashboard
+```
+
+### Option B: Run Locally (Development)
+
+If you prefer running services locally without Docker:
+
+```bash
+# Start PostgreSQL with Docker
+docker-compose up -d postgres
+
+# Wait for database to be healthy, then initialize
+python scripts/init_database.py
+
+# Start FastAPI backend
+uvicorn SPARC.api:app --host 0.0.0.0 --port 8000 --reload
+
+# For the React frontend (separate terminal)
+cd frontend
+npm install
+npm run dev
+```
+
+---
+
+## Step 5: Verify Deployment
+
+```bash
+# Check API health
+curl http://localhost:8000/health
+
+# Expected response:
+# {"status":"healthy","version":"0.1.0","timestamp":"..."}
+```
+
+Access the services:
+
+| Service | URL |
+|---------|-----|
+| REST API | http://localhost:8000 |
+| API Documentation (Swagger) | http://localhost:8000/docs |
+| Dashboard (Web UI) | http://localhost:8080 |
+
+---
+
+## Step 6: Using the Application
+
+### Via Dashboard (Web UI)
+
+1. Open http://localhost:8080
+2. Register a new account or login (default admin: `admin` / `admin`)
+3. Navigate to **"Analysis"** from the sidebar
+4. Enter a company name (e.g., "Intel")
+5. Click **"Analyze"**
+
+This will:
+- Query SerpAPI for recent patents
+- Download and parse patent PDFs
+- Send patent content to Claude for analysis
+- Store prompt/response in PostgreSQL (with caching)
+- Display results in the dashboard
+
+### Via REST API
+
+```bash
+# Analyze single company
+curl http://localhost:8000/analyze/Intel
+
+# Batch analyze multiple companies (synchronous)
+curl -X POST http://localhost:8000/analyze/batch \
+  -H "Content-Type: application/json" \
+  -d '{"companies": ["Intel", "AMD", "NVIDIA"], "max_workers": 3}'
+
+# Async batch (for large jobs)
+curl -X POST http://localhost:8000/analyze/batch/async \
+  -H "Content-Type: application/json" \
+  -d '{"companies": ["Intel", "AMD"]}'
+
+# Check job status
+curl http://localhost:8000/jobs/{job_id}
+
+# List all jobs
+curl http://localhost:8000/jobs
+```
+
+### Via Python
+
+```python
+from SPARC.analyzer import CompanyAnalyzer
+
+analyzer = CompanyAnalyzer()
+result = analyzer.analyze("Intel")
+print(result.analysis)
+```
+
+---
+
+## Step 7: View Stored Data
+
+```bash
+# View analytics (aggregated usage)
+python scripts/view_analytics.py
+
+# View stored messages
+python scripts/view_messages.py
+
+# Query database directly
+docker exec -it sparc-postgres psql -U postgres -d sparc -c \
+  "SELECT company_name, analysis_type, token_usage FROM llm_messages ORDER BY timestamp DESC LIMIT 10;"
+```
+
+---
+
+## Architecture Overview
+
+```
+┌──────────────┐    ┌──────────────┐    ┌──────────────┐
+│  Dashboard   │───▶│   FastAPI    │───▶│   Analyzer   │
+│  (8501)      │    │   (8000)     │    │              │
+└──────────────┘    └──────────────┘    └──────┬───────┘
+                                               │
+                    ┌──────────────────────────┼──────────────────────────┐
+                    │                          │                          │
+                    ▼                          ▼                          ▼
+           ┌──────────────┐           ┌──────────────┐           ┌──────────────┐
+           │   SerpAPI    │           │  OpenRouter  │           │  PostgreSQL  │
+           │ (Patents)    │           │  (Claude)    │           │  (Storage)   │
+           └──────────────┘           └──────────────┘           └──────────────┘
+```
+
+### Component Responsibilities
+
+| Component | Purpose |
+|-----------|---------|
+| **Dashboard** | React TypeScript web UI with authentication |
+| **FastAPI** | REST API with JWT authentication |
+| **Analyzer** | Orchestrates patent retrieval and LLM analysis |
+| **SerpAPI** | Retrieves patent data from Google Patents |
+| **OpenRouter** | Routes requests to Claude for AI analysis |
+| **PostgreSQL** | Stores prompts, responses, users, and cached results |
+
+---
+
+## Environment Variables Reference
+
+| Variable | Required | Default | Description |
+|----------|----------|---------|-------------|
+| `API_KEY` | Yes | - | SerpAPI key for patent search |
+| `OPENROUTER_API_KEY` | Yes | - | OpenRouter API key for Claude access |
+| `DATABASE_URL` | Yes | - | PostgreSQL connection string |
+| `USE_CACHE` | No | `true` | Check database for cached responses before API calls |
+| `JWT_SECRET` | Yes | - | Secret key for JWT authentication (change in production!) |
+
+### Database URL Format
+
+```
+postgresql://[user]:[password]@[host]:[port]/[database]
+```
+
+Example:
+```
+postgresql://postgres:postgres@localhost:5432/sparc
+```
+
+---
+
+## Docker Compose Services
+
+The `docker-compose.yml` includes all services needed for production:
+
+| Service | Container | Port | Description |
+|---------|-----------|------|-------------|
+| `postgres` | sparc-postgres | 5432 | PostgreSQL database |
+| `init-db` | sparc-init-db | - | One-time database initialization (seeds admin user) |
+| `api` | sparc-api | 8000 | FastAPI REST API with JWT auth (patent PDFs stored in `patent_data` volume) |
+| `dashboard` | sparc-dashboard | 8080 | React TypeScript web UI |
+
+### Common Docker Compose Commands
+
+```bash
+# Start all services
+docker-compose up -d
+
+# Start with rebuild (after code changes)
+docker-compose up -d --build
+
+# View logs
+docker-compose logs -f
+
+# View specific service logs
+docker-compose logs -f api
+docker-compose logs -f dashboard
+
+# Stop all services
+docker-compose down
+
+# Stop and remove volumes (WARNING: deletes data)
+docker-compose down -v
+
+# Restart a specific service
+docker-compose restart api
+```
+
+---
+
+## Patent PDF Storage
+
+The SPARC API downloads patent PDFs during analysis and stores them at `/app/patents` inside the container. These files are used for subsequent single-patent analysis requests and as a local cache to avoid re-downloading. If this directory is not persisted, all downloaded PDFs are lost when the container is recreated.
+
+### Docker Compose (default)
+
+The default `docker-compose.yml` declares a named volume called `patent_data` that is mounted at `/app/patents`:
+
+```yaml
+# In the api service:
+volumes:
+  - patent_data:/app/patents
+
+# At the top-level volumes section:
+volumes:
+  patent_data:
+```
+
+This means PDFs survive `docker compose down` and `docker compose up` cycles. To remove patent data intentionally, run:
+
+```bash
+docker compose down -v   # WARNING: also removes postgres_data
+# or selectively:
+docker volume rm sparc_patent_data
+```
+
+If you prefer a bind mount (e.g., for easy host-side access during development), replace the volume with:
+
+```yaml
+volumes:
+  - ./patents:/app/patents
+```
+
+### Kubernetes
+
+For Kubernetes deployments, create a PersistentVolumeClaim and mount it into the API pod:
+
+```yaml
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+  name: sparc-patent-data
+spec:
+  accessModes:
+    - ReadWriteOnce
+  resources:
+    requests:
+      storage: 5Gi
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: sparc-api
+spec:
+  template:
+    spec:
+      containers:
+        - name: api
+          volumeMounts:
+            - name: patent-data
+              mountPath: /app/patents
+      volumes:
+        - name: patent-data
+          persistentVolumeClaim:
+            claimName: sparc-patent-data
+```
+
+Adjust the storage size based on expected patent volume. Each patent PDF is typically 1-5 MB.
+
+### S3 Object Storage (alternative)
+
+For production deployments that need shared or highly durable storage, set `STORAGE_BACKEND=s3` in your `.env` file. This stores patent PDFs in an S3-compatible bucket (AWS S3 or MinIO) instead of the local filesystem, eliminating the need for a persistent volume. See the S3/MinIO section in `.env.example` for configuration details.
+
+---
+
+## Troubleshooting
+
+### Database Connection Issues
+
+```bash
+# Check if postgres is running
+docker-compose ps
+
+# Check postgres logs
+docker-compose logs postgres
+
+# Test database connection
+docker exec -it sparc-postgres psql -U postgres -d sparc -c "SELECT 1;"
+```
+
+### API Key Issues
+
+```bash
+# Verify environment variables are set
+echo $API_KEY
+echo $OPENROUTER_API_KEY
+
+# Test SerpAPI directly
+curl "https://serpapi.com/search?engine=google_patents&q=Intel&api_key=$API_KEY"
+```
+
+### Port Conflicts
+
+If ports 8000, 8501, or 5432 are in use:
+
+```bash
+# Find what's using the port
+lsof -i :8000
+
+# Or change ports in docker-compose.yml
+ports:
+  - "8080:8000"  # Use 8080 instead of 8000
+```
+
+### Container Issues
+
+```bash
+# Rebuild containers after code changes
+docker-compose build --no-cache
+
+# Remove all containers and start fresh
+docker-compose down
+docker-compose up -d --build
+```
+
+### Viewing Application Logs
+
+```bash
+# All services
+docker-compose logs -f
+
+# Specific service
+docker-compose logs -f api
+docker-compose logs -f dashboard
+```
+
+---
+
+## Quick Reference
+
+```bash
+# Docker setup (recommended)
+cp .env.example .env
+# Edit .env with API keys
+docker-compose up -d
+
+# Local development setup
+cp .env.example .env
+# Edit .env with API keys
+docker-compose up -d postgres
+python scripts/init_database.py
+uvicorn SPARC.api:app --reload &
+cd frontend && npm install && npm run dev &
+
+# Check status
+curl http://localhost:8000/health
+open http://localhost:8080
+
+# View data
+python scripts/view_analytics.py
+python scripts/view_messages.py
+```

flake.nix

@@ -20,6 +20,14 @@
           packages = [
             python
             pkgs.python311Packages.virtualenv  # gives `virtualenv` tool
+            pkgs.zlib
+            pkgs.stdenv.cc.cc.lib
+          ];
+
+          # Required for numpy and other C extension packages
+          LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath [
+            pkgs.zlib
+            pkgs.stdenv.cc.cc.lib
           ];
 
           shellHook = ''
@@ -48,8 +56,8 @@
             fi
 
             # Prompt tweak so you can see when venv is active
-            export PS1="(SPARC-venv) $PS1"
+            export NIX_PROJECT_SHELL="SPARC"
           '';
         };
       });
-}
+}

frontend/.gitignore

@@ -0,0 +1,22 @@
+# Dependencies
+node_modules/
+
+# Build output
+dist/
+
+# Local env files
+.env.local
+.env.*.local
+
+# Editor directories
+.vscode/
+.idea/
+
+# OS files
+.DS_Store
+Thumbs.db
+
+# Debug logs
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*

frontend/Dockerfile

@@ -0,0 +1,32 @@
+# Build stage
+FROM node:20-alpine AS build
+
+WORKDIR /app
+
+# Copy package files
+COPY package.json package-lock.json* ./
+
+# Install dependencies
+RUN npm install
+
+# Copy source files
+COPY . .
+
+# Build the application
+RUN npm run build
+
+# Production stage
+FROM nginx:alpine
+
+# Copy built files
+COPY --from=build /app/dist /usr/share/nginx/html
+
+# Copy nginx template (processed at startup with envsubst)
+COPY nginx.conf.template /etc/nginx/templates/default.conf.template
+
+# Default API URL (override with -e API_URL=...)
+ENV API_URL=http://api:8000/
+
+EXPOSE 80
+
+CMD ["nginx", "-g", "daemon off;"]

frontend/index.html

@@ -0,0 +1,22 @@
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <link rel="icon" type="image/svg+xml" href="/vite.svg" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>SPARC Dashboard</title>
+  </head>
+  <body>
+    <script>
+      // Prevent FOUC: apply saved theme before first render
+      (function() {
+        var theme = localStorage.getItem('theme');
+        if (theme === 'dark' || (!theme && window.matchMedia('(prefers-color-scheme: dark)').matches)) {
+          document.documentElement.classList.add('dark');
+        }
+      })();
+    </script>
+    <div id="root"></div>
+    <script type="module" src="/src/main.tsx"></script>
+  </body>
+</html>

frontend/nginx.conf.template

@@ -0,0 +1,34 @@
+server {
+    listen 80;
+    server_name localhost;
+    root /usr/share/nginx/html;
+    index index.html;
+
+    # Gzip compression
+    gzip on;
+    gzip_types text/plain text/css application/json application/javascript text/xml application/xml application/xml+rss text/javascript;
+
+    # Handle React Router (SPA)
+    location / {
+        try_files $uri $uri/ /index.html;
+    }
+
+    # Proxy API requests to backend
+    location /api/ {
+        proxy_pass ${API_URL};
+        proxy_http_version 1.1;
+        proxy_set_header Upgrade $http_upgrade;
+        proxy_set_header Connection 'upgrade';
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+        proxy_cache_bypass $http_upgrade;
+    }
+
+    # Cache static assets
+    location ~* \.(js|css|png|jpg|jpeg|gif|ico|svg|woff|woff2)$ {
+        expires 1y;
+        add_header Cache-Control "public, immutable";
+    }
+}

frontend/package.json

@@ -0,0 +1,41 @@
+{
+  "name": "sparc-dashboard",
+  "private": true,
+  "version": "1.0.0",
+  "type": "module",
+  "scripts": {
+    "dev": "vite",
+    "build": "tsc -b && vite build",
+    "lint": "eslint .",
+    "generate": "openapi-typescript http://localhost:8000/api/openapi.json -o src/api/schema.d.ts",
+    "generate:local": "openapi-typescript src/api/openapi.json -o src/api/schema.d.ts",
+    "typecheck": "tsc --noEmit",
+    "preview": "vite preview"
+  },
+  "dependencies": {
+    "@tanstack/react-query": "^5.51.0",
+    "axios": "^1.7.2",
+    "lucide-react": "^1.7.0",
+    "react": "^18.3.1",
+    "react-dom": "^18.3.1",
+    "react-router-dom": "^6.24.0",
+    "recharts": "^2.12.7"
+  },
+  "devDependencies": {
+    "@eslint/js": "^9.6.0",
+    "@types/react": "^18.3.3",
+    "@types/react-dom": "^18.3.0",
+    "@vitejs/plugin-react": "^4.3.1",
+    "autoprefixer": "^10.4.19",
+    "eslint": "^9.6.0",
+    "eslint-plugin-react-hooks": "^5.1.0",
+    "eslint-plugin-react-refresh": "^0.4.7",
+    "globals": "^15.8.0",
+    "postcss": "^8.4.39",
+    "tailwindcss": "^3.4.4",
+    "openapi-typescript": "^7.0.0",
+    "typescript": "~5.5.3",
+    "typescript-eslint": "^8.0.0",
+    "vite": "^5.3.3"
+  }
+}

frontend/postcss.config.js

@@ -0,0 +1,6 @@
+export default {
+  plugins: {
+    tailwindcss: {},
+    autoprefixer: {},
+  },
+}

frontend/src/App.tsx

@@ -0,0 +1,72 @@
+import { BrowserRouter, Routes, Route, Navigate } from 'react-router-dom';
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query';
+import { AuthProvider } from './context/AuthContext';
+import { ThemeProvider } from './context/ThemeContext';
+import { Layout } from './components/Layout';
+import { ProtectedRoute } from './components/ProtectedRoute';
+import { Login } from './pages/Login';
+import { Register } from './pages/Register';
+import { Analysis } from './pages/Analysis';
+import { Batch } from './pages/Batch';
+import { AnalyticsPage } from './pages/Analytics';
+import { About } from './pages/About';
+import { AdminUsers } from './pages/AdminUsers';
+import { Compare } from './pages/Compare';
+
+const queryClient = new QueryClient({
+  defaultOptions: {
+    queries: {
+      staleTime: 1000 * 60 * 5, // 5 minutes
+      retry: 1,
+    },
+  },
+});
+
+function App() {
+  return (
+    <ThemeProvider>
+    <QueryClientProvider client={queryClient}>
+      <AuthProvider>
+        <BrowserRouter>
+          <Routes>
+            {/* Public routes */}
+            <Route path="/login" element={<Login />} />
+            <Route path="/register" element={<Register />} />
+
+            {/* Protected routes */}
+            <Route
+              element={
+                <ProtectedRoute>
+                  <Layout />
+                </ProtectedRoute>
+              }
+            >
+              <Route path="/analysis" element={<Analysis />} />
+              <Route path="/batch" element={<Batch />} />
+              <Route path="/analytics" element={<AnalyticsPage />} />
+              <Route path="/compare" element={<Compare />} />
+              <Route path="/about" element={<About />} />
+
+              {/* Admin routes */}
+              <Route
+                path="/admin/users"
+                element={
+                  <ProtectedRoute requireAdmin>
+                    <AdminUsers />
+                  </ProtectedRoute>
+                }
+              />
+            </Route>
+
+            {/* Default redirect */}
+            <Route path="/" element={<Navigate to="/analysis" replace />} />
+            <Route path="*" element={<Navigate to="/analysis" replace />} />
+          </Routes>
+        </BrowserRouter>
+      </AuthProvider>
+    </QueryClientProvider>
+    </ThemeProvider>
+  );
+}
+
+export default App;

frontend/src/api/client.ts

@@ -0,0 +1,221 @@
+import axios, { AxiosError, InternalAxiosRequestConfig } from 'axios';
+import type { TokenResponse, User, CompanyAnalysis, BatchAnalysisResult, JobStatus, Analytics } from '../types';
+
+const API_BASE_URL = import.meta.env.VITE_API_URL || '/api';
+
+const api = axios.create({
+  baseURL: API_BASE_URL,
+  headers: {
+    'Content-Type': 'application/json',
+  },
+});
+
+// Token management
+let accessToken: string | null = localStorage.getItem('access_token');
+let refreshToken: string | null = localStorage.getItem('refresh_token');
+
+export const setTokens = (tokens: TokenResponse) => {
+  accessToken = tokens.access_token;
+  refreshToken = tokens.refresh_token;
+  localStorage.setItem('access_token', tokens.access_token);
+  localStorage.setItem('refresh_token', tokens.refresh_token);
+};
+
+export const clearTokens = () => {
+  accessToken = null;
+  refreshToken = null;
+  localStorage.removeItem('access_token');
+  localStorage.removeItem('refresh_token');
+};
+
+export const getAccessToken = () => accessToken;
+
+// Request interceptor to add auth header
+api.interceptors.request.use((config: InternalAxiosRequestConfig) => {
+  if (accessToken) {
+    config.headers.Authorization = `Bearer ${accessToken}`;
+  }
+  return config;
+});
+
+// Response interceptor to handle token refresh
+api.interceptors.response.use(
+  (response) => response,
+  async (error: AxiosError) => {
+    const originalRequest = error.config as InternalAxiosRequestConfig & { _retry?: boolean };
+
+    if (error.response?.status === 401 && !originalRequest._retry && refreshToken) {
+      originalRequest._retry = true;
+
+      try {
+        const response = await axios.post<TokenResponse>(`${API_BASE_URL}/auth/refresh`, {
+          refresh_token: refreshToken,
+        });
+
+        setTokens(response.data);
+        originalRequest.headers.Authorization = `Bearer ${response.data.access_token}`;
+
+        return api(originalRequest);
+      } catch {
+        clearTokens();
+        window.location.href = '/login';
+      }
+    }
+
+    return Promise.reject(error);
+  }
+);
+
+// Auth API
+export const authApi = {
+  register: async (email: string, password: string): Promise<User> => {
+    const response = await api.post<User>('/auth/register', { email, password });
+    return response.data;
+  },
+
+  login: async (email: string, password: string): Promise<TokenResponse> => {
+    const response = await api.post<TokenResponse>('/auth/login', { email, password });
+    setTokens(response.data);
+    return response.data;
+  },
+
+  getMe: async (): Promise<User> => {
+    const response = await api.get<User>('/auth/me');
+    return response.data;
+  },
+
+  logout: () => {
+    clearTokens();
+  },
+};
+
+// Model types
+export interface ModelInfo {
+  id: string;
+  name: string;
+  provider: string;
+}
+
+export interface ModelsResponse {
+  models: ModelInfo[];
+  default: string;
+}
+
+// Analysis API
+export const analysisApi = {
+  analyzeCompany: async (companyName: string, model?: string): Promise<CompanyAnalysis> => {
+    const params = new URLSearchParams();
+    if (model) params.append('model', model);
+    const qs = params.toString();
+    const response = await api.get<CompanyAnalysis>(
+      `/analyze/${encodeURIComponent(companyName)}${qs ? `?${qs}` : ''}`
+    );
+    return response.data;
+  },
+
+  analyzeBatch: async (companies: string[], maxWorkers = 3, model?: string): Promise<BatchAnalysisResult> => {
+    const response = await api.post<BatchAnalysisResult>('/analyze/batch', {
+      companies,
+      max_workers: maxWorkers,
+      ...(model ? { model } : {}),
+    });
+    return response.data;
+  },
+
+  analyzeBatchAsync: async (companies: string[], maxWorkers = 3, model?: string): Promise<JobStatus> => {
+    const response = await api.post<JobStatus>('/analyze/batch/async', {
+      companies,
+      max_workers: maxWorkers,
+      ...(model ? { model } : {}),
+    });
+    return response.data;
+  },
+
+  listModels: async (): Promise<ModelsResponse> => {
+    const response = await api.get<ModelsResponse>('/models');
+    return response.data;
+  },
+
+  getJobStatus: async (jobId: string): Promise<JobStatus> => {
+    const response = await api.get<JobStatus>(`/jobs/${jobId}`);
+    return response.data;
+  },
+
+  listJobs: async (status?: string, limit = 10): Promise<JobStatus[]> => {
+    const params = new URLSearchParams();
+    if (status) params.append('status', status);
+    params.append('limit', limit.toString());
+    const response = await api.get<JobStatus[]>(`/jobs?${params}`);
+    return response.data;
+  },
+};
+
+// Export API
+export const exportApi = {
+  exportCsv: async (companyName: string): Promise<void> => {
+    const response = await api.get(`/export/${encodeURIComponent(companyName)}`, {
+      responseType: 'blob',
+    });
+    const url = window.URL.createObjectURL(new Blob([response.data]));
+    const link = document.createElement('a');
+    link.href = url;
+    link.setAttribute('download', `sparc_${companyName.toLowerCase().replace(/\s+/g, '_')}_export.csv`);
+    document.body.appendChild(link);
+    link.click();
+    link.remove();
+    window.URL.revokeObjectURL(url);
+  },
+  exportPdf: async (companyName: string): Promise<void> => {
+    const response = await api.get(`/export/${encodeURIComponent(companyName)}/pdf`, {
+      responseType: 'blob',
+    });
+    const safeName = companyName.toLowerCase().replace(/\s+/g, '_');
+    const date = new Date().toISOString().split('T')[0];
+    const url = window.URL.createObjectURL(new Blob([response.data], { type: 'application/pdf' }));
+    const link = document.createElement('a');
+    link.href = url;
+    link.setAttribute('download', `${safeName}-analysis-${date}.pdf`);
+    document.body.appendChild(link);
+    link.click();
+    link.remove();
+    window.URL.revokeObjectURL(url);
+  },
+};
+
+// Analytics API
+export interface TrendData {
+  by_month: Array<{ month: string; company_name: string; count: number }>;
+  by_type_over_time: Array<{ month: string; analysis_type: string; count: number }>;
+  period_days: number;
+}
+
+export const analyticsApi = {
+  getAnalytics: async (days = 30): Promise<Analytics> => {
+    const response = await api.get<Analytics>(`/analytics?days=${days}`);
+    return response.data;
+  },
+
+  getTrends: async (days = 90): Promise<TrendData> => {
+    const response = await api.get<TrendData>(`/analytics/trends?days=${days}`);
+    return response.data;
+  },
+};
+
+// Admin API
+export const adminApi = {
+  listUsers: async (limit = 100, offset = 0): Promise<User[]> => {
+    const response = await api.get<User[]>(`/admin/users?limit=${limit}&offset=${offset}`);
+    return response.data;
+  },
+
+  updateUserRole: async (userId: number, role: 'admin' | 'user'): Promise<User> => {
+    const response = await api.patch<User>(`/admin/users/${userId}/role`, { role });
+    return response.data;
+  },
+
+  deleteUser: async (userId: number): Promise<void> => {
+    await api.delete(`/admin/users/${userId}`);
+  },
+};
+
+export default api;

frontend/src/api/schema.d.ts

@@ -0,0 +1,975 @@
+/**
+ * This file was auto-generated by openapi-typescript.
+ * Do not make direct changes to the file.
+ */
+
+export interface paths {
+    "/auth/register": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Register
+         * @description Register a new user.
+         *
+         *     The first registered user automatically becomes an admin.
+         */
+        post: operations["register_auth_register_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/auth/login": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Login
+         * @description Authenticate user and return JWT tokens.
+         */
+        post: operations["login_auth_login_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/auth/refresh": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Refresh Token
+         * @description Refresh access token using refresh token.
+         */
+        post: operations["refresh_token_auth_refresh_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/auth/me": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Get Me
+         * @description Get current authenticated user.
+         */
+        get: operations["get_me_auth_me_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/admin/users": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * List Users
+         * @description List all users (admin only).
+         */
+        get: operations["list_users_admin_users_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/admin/users/{user_id}/role": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        /**
+         * Update User Role
+         * @description Update a user's role (admin only).
+         */
+        patch: operations["update_user_role_admin_users__user_id__role_patch"];
+        trace?: never;
+    };
+    "/admin/users/{user_id}": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        post?: never;
+        /**
+         * Delete User
+         * @description Delete a user (admin only).
+         */
+        delete: operations["delete_user_admin_users__user_id__delete"];
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/analytics": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Get Analytics
+         * @description Get analytics data (authenticated users only).
+         */
+        get: operations["get_analytics_analytics_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/health": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Health Check
+         * @description Check API health status.
+         */
+        get: operations["health_check_health_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/analyze/{company_name}": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Analyze Company
+         * @description Analyze a single company's patent portfolio.
+         *
+         *     This endpoint retrieves recent patents for the specified company,
+         *     parses them, and uses AI to generate a comprehensive analysis.
+         *
+         *     Args:
+         *         company_name: Name of the company to analyze (e.g., "nvidia", "intel")
+         *
+         *     Returns:
+         *         Analysis results including patent count, AI insights, and success status
+         */
+        get: operations["analyze_company_analyze__company_name__get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/analyze/batch": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Analyze Companies Batch
+         * @description Analyze multiple companies' patent portfolios.
+         *
+         *     Processes companies concurrently for improved performance.
+         *     Limited to 20 companies per request.
+         *
+         *     Args:
+         *         request: List of company names and optional worker count
+         *
+         *     Returns:
+         *         Batch results with individual company analyses and summary statistics
+         */
+        post: operations["analyze_companies_batch_analyze_batch_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/analyze/batch/async": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Analyze Companies Async
+         * @description Start an asynchronous batch analysis job.
+         *
+         *     Returns immediately with a job ID that can be used to poll for status.
+         *     Useful for large batch analyses that may take a long time.
+         *
+         *     Args:
+         *         request: List of company names and optional worker count
+         *
+         *     Returns:
+         *         Job status with job_id for polling
+         */
+        post: operations["analyze_companies_async_analyze_batch_async_post"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/jobs/{job_id}": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Get Job Status
+         * @description Get the status of a background analysis job.
+         *
+         *     Args:
+         *         job_id: The job ID returned from the async batch endpoint
+         *
+         *     Returns:
+         *         Current job status including progress and results when complete
+         */
+        get: operations["get_job_status_jobs__job_id__get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/jobs": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * List Jobs
+         * @description List all analysis jobs.
+         *
+         *     Args:
+         *         status: Optional filter by job status
+         *         limit: Maximum number of jobs to return (default 10, max 100)
+         *
+         *     Returns:
+         *         List of job statuses
+         */
+        get: operations["list_jobs_jobs_get"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+}
+export type webhooks = Record<string, never>;
+export interface components {
+    schemas: {
+        /**
+         * AnalyticsResponse
+         * @description Analytics response model.
+         */
+        AnalyticsResponse: {
+            /** Total Messages */
+            total_messages: number;
+            /** By Company */
+            by_company: {
+                [key: string]: unknown;
+            }[];
+            /** By Type */
+            by_type: {
+                [key: string]: unknown;
+            }[];
+            /** Period Days */
+            period_days: number;
+        };
+        /**
+         * BatchAnalysisRequest
+         * @description Request model for batch company analysis.
+         */
+        BatchAnalysisRequest: {
+            /**
+             * Companies
+             * @description List of company names to analyze
+             */
+            companies: string[];
+            /**
+             * Max Workers
+             * @description Max concurrent analyses
+             * @default 3
+             */
+            max_workers: number;
+        };
+        /**
+         * BatchAnalysisResponse
+         * @description Response model for batch company analysis.
+         */
+        BatchAnalysisResponse: {
+            /** Results */
+            results: components["schemas"]["CompanyAnalysisResponse"][];
+            /** Total Companies */
+            total_companies: number;
+            /** Successful */
+            successful: number;
+            /** Failed */
+            failed: number;
+            /**
+             * Timestamp
+             * Format: date-time
+             */
+            timestamp: string;
+        };
+        /**
+         * CompanyAnalysisResponse
+         * @description Response model for single company analysis.
+         */
+        CompanyAnalysisResponse: {
+            /** Company Name */
+            company_name: string;
+            /** Analysis */
+            analysis: string;
+            /** Patent Count */
+            patent_count: number;
+            /** Success */
+            success: boolean;
+            /** Error */
+            error?: string | null;
+            /**
+             * Timestamp
+             * Format: date-time
+             */
+            timestamp: string;
+        };
+        /** HTTPValidationError */
+        HTTPValidationError: {
+            /** Detail */
+            detail?: components["schemas"]["ValidationError"][];
+        };
+        /**
+         * HealthResponse
+         * @description Health check response.
+         */
+        HealthResponse: {
+            /** Status */
+            status: string;
+            /** Version */
+            version: string;
+            /**
+             * Timestamp
+             * Format: date-time
+             */
+            timestamp: string;
+        };
+        /**
+         * JobStatus
+         * @description Status of a background analysis job.
+         */
+        JobStatus: {
+            /** Job Id */
+            job_id: string;
+            /** Status */
+            status: string;
+            /** Progress */
+            progress: number;
+            /** Total Companies */
+            total_companies: number;
+            /** Completed Companies */
+            completed_companies: number;
+            result?: components["schemas"]["BatchAnalysisResponse"] | null;
+            /** Error */
+            error?: string | null;
+        };
+        /**
+         * LoginRequest
+         * @description User login request.
+         */
+        LoginRequest: {
+            /**
+             * Email
+             * Format: email
+             */
+            email: string;
+            /** Password */
+            password: string;
+        };
+        /**
+         * RefreshRequest
+         * @description Token refresh request.
+         */
+        RefreshRequest: {
+            /** Refresh Token */
+            refresh_token: string;
+        };
+        /**
+         * RegisterRequest
+         * @description User registration request.
+         */
+        RegisterRequest: {
+            /**
+             * Email
+             * Format: email
+             */
+            email: string;
+            /**
+             * Password
+             * @description Password (min 8 characters)
+             */
+            password: string;
+        };
+        /**
+         * TokenResponse
+         * @description Token response model.
+         */
+        TokenResponse: {
+            /** Access Token */
+            access_token: string;
+            /** Refresh Token */
+            refresh_token: string;
+            /**
+             * Token Type
+             * @default bearer
+             */
+            token_type: string;
+        };
+        /**
+         * UpdateRoleRequest
+         * @description Update user role request.
+         */
+        UpdateRoleRequest: {
+            /** Role */
+            role: string;
+        };
+        /**
+         * UserResponse
+         * @description User response model.
+         */
+        UserResponse: {
+            /** Id */
+            id: number;
+            /** Email */
+            email: string;
+            /** Role */
+            role: string;
+            /**
+             * Created At
+             * Format: date-time
+             */
+            created_at: string;
+        };
+        /** ValidationError */
+        ValidationError: {
+            /** Location */
+            loc: (string | number)[];
+            /** Message */
+            msg: string;
+            /** Error Type */
+            type: string;
+            /** Input */
+            input?: unknown;
+            /** Context */
+            ctx?: Record<string, never>;
+        };
+    };
+    responses: never;
+    parameters: never;
+    requestBodies: never;
+    headers: never;
+    pathItems: never;
+}
+export type $defs = Record<string, never>;
+export interface operations {
+    register_auth_register_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["RegisterRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["UserResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    login_auth_login_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["LoginRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["TokenResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    refresh_token_auth_refresh_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["RefreshRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["TokenResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    get_me_auth_me_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["UserResponse"];
+                };
+            };
+        };
+    };
+    list_users_admin_users_get: {
+        parameters: {
+            query?: {
+                limit?: number;
+                offset?: number;
+            };
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["UserResponse"][];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    update_user_role_admin_users__user_id__role_patch: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                user_id: number;
+            };
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["UpdateRoleRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["UserResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    delete_user_admin_users__user_id__delete: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                user_id: number;
+            };
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": unknown;
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    get_analytics_analytics_get: {
+        parameters: {
+            query?: {
+                days?: number;
+            };
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AnalyticsResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    health_check_health_get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HealthResponse"];
+                };
+            };
+        };
+    };
+    analyze_company_analyze__company_name__get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                company_name: string;
+            };
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["CompanyAnalysisResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    analyze_companies_batch_analyze_batch_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["BatchAnalysisRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["BatchAnalysisResponse"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    analyze_companies_async_analyze_batch_async_post: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["BatchAnalysisRequest"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["JobStatus"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    get_job_status_jobs__job_id__get: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                job_id: string;
+            };
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["JobStatus"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    list_jobs_jobs_get: {
+        parameters: {
+            query?: {
+                /** @description Filter by status: pending, running, completed, failed */
+                status?: string | null;
+                limit?: number;
+            };
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["JobStatus"][];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+}

frontend/src/components/Layout.tsx

@@ -0,0 +1,118 @@
+import { Outlet, NavLink, useNavigate } from 'react-router-dom';
+import { useAuth } from '../context/AuthContext';
+import { useTheme } from '../context/ThemeContext';
+import { Search, Layers, BarChart3, Info, Users, LogOut, GitCompareArrows, Sun, Moon } from 'lucide-react';
+
+export function Layout() {
+  const { user, isAdmin, logout } = useAuth();
+  const { theme, toggleTheme } = useTheme();
+  const navigate = useNavigate();
+
+  const handleLogout = () => {
+    logout();
+    navigate('/login');
+  };
+
+  const navItems = [
+    { to: '/analysis', icon: Search, label: 'Analysis' },
+    { to: '/batch', icon: Layers, label: 'Batch' },
+    { to: '/analytics', icon: BarChart3, label: 'Analytics' },
+    { to: '/compare', icon: GitCompareArrows, label: 'Compare' },
+    { to: '/about', icon: Info, label: 'About' },
+  ];
+
+  if (isAdmin) {
+    navItems.push({ to: '/admin/users', icon: Users, label: 'Users' });
+  }
+
+  return (
+    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-slate-100 dark:to-indigo-950">
+      {/* Header */}
+      <header className="bg-bg-card/80 backdrop-blur-lg border-b border-primary/20">
+        <div className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8">
+          <div className="flex items-center justify-between h-16">
+            {/* Brand */}
+            <div className="flex items-center gap-3">
+              <span className="text-2xl">⚡</span>
+              <div>
+                <h1 className="text-xl font-bold bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent">
+                  SPARC
+                </h1>
+                <span className="text-xs text-text-secondary uppercase tracking-wider">
+                  Semiconductor Patent Analytics
+                </span>
+              </div>
+            </div>
+
+            {/* Navigation */}
+            <nav className="hidden md:flex items-center gap-1 bg-bg-card/60 rounded-xl p-1 border border-primary/15">
+              {navItems.map(({ to, icon: Icon, label }) => (
+                <NavLink
+                  key={to}
+                  to={to}
+                  className={({ isActive }) =>
+                    `flex items-center gap-2 px-4 py-2 rounded-lg text-sm font-medium transition-all ${
+                      isActive
+                        ? 'bg-gradient-to-r from-primary to-primary-dark text-white'
+                        : 'text-text-secondary hover:text-text-primary hover:bg-bg-card-hover'
+                    }`
+                  }
+                >
+                  <Icon size={16} />
+                  {label}
+                </NavLink>
+              ))}
+            </nav>
+
+            {/* User menu */}
+            <div className="flex items-center gap-4">
+              <button
+                onClick={toggleTheme}
+                className="p-2 rounded-lg text-text-secondary hover:text-text-primary hover:bg-bg-card-hover transition-all"
+                aria-label={theme === 'dark' ? 'Switch to light mode' : 'Switch to dark mode'}
+              >
+                {theme === 'dark' ? <Sun size={18} /> : <Moon size={18} />}
+              </button>
+              <div className="text-right hidden sm:block">
+                <div className="text-sm font-medium text-text-primary">{user?.email}</div>
+                <div className="text-xs text-text-secondary capitalize">{user?.role}</div>
+              </div>
+              <button
+                onClick={handleLogout}
+                className="flex items-center gap-2 px-3 py-2 rounded-lg text-text-secondary hover:text-error hover:bg-error/10 transition-all"
+              >
+                <LogOut size={18} />
+                <span className="hidden sm:inline">Logout</span>
+              </button>
+            </div>
+          </div>
+        </div>
+      </header>
+
+      {/* Mobile Navigation */}
+      <nav className="md:hidden fixed bottom-0 left-0 right-0 bg-bg-card/95 backdrop-blur-lg border-t border-primary/20 z-50">
+        <div className="flex justify-around py-2">
+          {navItems.map(({ to, icon: Icon, label }) => (
+            <NavLink
+              key={to}
+              to={to}
+              className={({ isActive }) =>
+                `flex flex-col items-center gap-1 px-3 py-2 rounded-lg text-xs font-medium transition-all ${
+                  isActive ? 'text-primary' : 'text-text-secondary'
+                }`
+              }
+            >
+              <Icon size={20} />
+              {label}
+            </NavLink>
+          ))}
+        </div>
+      </nav>
+
+      {/* Main content */}
+      <main className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-8 pb-24 md:pb-8">
+        <Outlet />
+      </main>
+    </div>
+  );
+}

frontend/src/components/ProtectedRoute.tsx

@@ -0,0 +1,30 @@
+import { Navigate, useLocation } from 'react-router-dom';
+import { useAuth } from '../context/AuthContext';
+
+interface ProtectedRouteProps {
+  children: React.ReactNode;
+  requireAdmin?: boolean;
+}
+
+export function ProtectedRoute({ children, requireAdmin = false }: ProtectedRouteProps) {
+  const { isAuthenticated, isAdmin, isLoading } = useAuth();
+  const location = useLocation();
+
+  if (isLoading) {
+    return (
+      <div className="min-h-screen bg-gradient-to-br from-bg-dark to-slate-100 dark:to-indigo-950 flex items-center justify-center">
+        <div className="animate-spin rounded-full h-12 w-12 border-t-2 border-b-2 border-primary"></div>
+      </div>
+    );
+  }
+
+  if (!isAuthenticated) {
+    return <Navigate to="/login" state={{ from: location }} replace />;
+  }
+
+  if (requireAdmin && !isAdmin) {
+    return <Navigate to="/analysis" replace />;
+  }
+
+  return <>{children}</>;
+}

frontend/src/context/AuthContext.tsx

@@ -0,0 +1,81 @@
+import { createContext, useContext, useState, useEffect, ReactNode } from 'react';
+import { authApi, getAccessToken } from '../api/client';
+import type { User } from '../types';
+
+interface AuthContextType {
+  user: User | null;
+  isLoading: boolean;
+  isAuthenticated: boolean;
+  isAdmin: boolean;
+  login: (email: string, password: string) => Promise<void>;
+  register: (email: string, password: string) => Promise<void>;
+  logout: () => void;
+  refreshUser: () => Promise<void>;
+}
+
+const AuthContext = createContext<AuthContextType | undefined>(undefined);
+
+export function AuthProvider({ children }: { children: ReactNode }) {
+  const [user, setUser] = useState<User | null>(null);
+  const [isLoading, setIsLoading] = useState(true);
+
+  const refreshUser = async () => {
+    try {
+      const userData = await authApi.getMe();
+      setUser(userData);
+    } catch {
+      setUser(null);
+    }
+  };
+
+  useEffect(() => {
+    const initAuth = async () => {
+      if (getAccessToken()) {
+        await refreshUser();
+      }
+      setIsLoading(false);
+    };
+    initAuth();
+  }, []);
+
+  const login = async (email: string, password: string) => {
+    await authApi.login(email, password);
+    await refreshUser();
+  };
+
+  const register = async (email: string, password: string) => {
+    await authApi.register(email, password);
+    await authApi.login(email, password);
+    await refreshUser();
+  };
+
+  const logout = () => {
+    authApi.logout();
+    setUser(null);
+  };
+
+  return (
+    <AuthContext.Provider
+      value={{
+        user,
+        isLoading,
+        isAuthenticated: !!user,
+        isAdmin: user?.role === 'admin',
+        login,
+        register,
+        logout,
+        refreshUser,
+      }}
+    >
+      {children}
+    </AuthContext.Provider>
+  );
+}
+
+export function useAuth() {
+  const context = useContext(AuthContext);
+  if (context === undefined) {
+    throw new Error('useAuth must be used within an AuthProvider');
+  }
+  return context;
+}

frontend/src/context/ThemeContext.tsx

@@ -0,0 +1,48 @@
+import { createContext, useContext, useEffect, useState } from 'react';
+
+type Theme = 'light' | 'dark';
+
+interface ThemeContextType {
+  theme: Theme;
+  toggleTheme: () => void;
+}
+
+const ThemeContext = createContext<ThemeContextType | undefined>(undefined);
+
+function getInitialTheme(): Theme {
+  const stored = localStorage.getItem('theme');
+  if (stored === 'light' || stored === 'dark') return stored;
+  return window.matchMedia('(prefers-color-scheme: dark)').matches ? 'dark' : 'light';
+}
+
+export function ThemeProvider({ children }: { children: React.ReactNode }) {
+  const [theme, setTheme] = useState<Theme>(getInitialTheme);
+
+  useEffect(() => {
+    const root = document.documentElement;
+    if (theme === 'dark') {
+      root.classList.add('dark');
+    } else {
+      root.classList.remove('dark');
+    }
+    localStorage.setItem('theme', theme);
+  }, [theme]);
+
+  const toggleTheme = () => {
+    setTheme((prev) => (prev === 'dark' ? 'light' : 'dark'));
+  };
+
+  return (
+    <ThemeContext.Provider value={{ theme, toggleTheme }}>
+      {children}
+    </ThemeContext.Provider>
+  );
+}
+
+export function useTheme() {
+  const context = useContext(ThemeContext);
+  if (!context) {
+    throw new Error('useTheme must be used within a ThemeProvider');
+  }
+  return context;
+}

frontend/src/context/useChartTheme.ts

@@ -0,0 +1,41 @@
+import { useTheme } from './ThemeContext';
+
+/**
+ * Returns theme-aware color values for recharts components.
+ *
+ * Recharts accepts only raw color strings (not CSS variables),
+ * so this hook bridges the Tailwind/CSS-variable theme system
+ * to the imperative recharts API.
+ */
+export function useChartTheme() {
+  const { theme } = useTheme();
+  const isDark = theme === 'dark';
+
+  return {
+    /** Axis tick and grid line stroke color */
+    axisStroke: isDark ? '#94a3b8' : '#64748b',
+    /** Tooltip container background */
+    tooltipBg: isDark ? '#1e293b' : '#ffffff',
+    /** Tooltip container border */
+    tooltipBorder: isDark
+      ? '1px solid rgba(99, 102, 241, 0.3)'
+      : '1px solid rgba(99, 102, 241, 0.2)',
+    /** Tooltip label text color */
+    tooltipLabelColor: isDark ? '#f8fafc' : '#0f172a',
+    /** Tooltip item text color */
+    tooltipItemColor: isDark ? '#e2e8f0' : '#334155',
+    /** Convenience: full contentStyle object for recharts Tooltip */
+    tooltipContentStyle: {
+      backgroundColor: isDark ? '#1e293b' : '#ffffff',
+      border: isDark
+        ? '1px solid rgba(99, 102, 241, 0.3)'
+        : '1px solid rgba(99, 102, 241, 0.2)',
+      borderRadius: '8px',
+      color: isDark ? '#f8fafc' : '#0f172a',
+    },
+    /** Convenience: labelStyle for recharts Tooltip */
+    tooltipLabelStyle: {
+      color: isDark ? '#f8fafc' : '#0f172a',
+    },
+  };
+}

frontend/src/index.css

@@ -0,0 +1,54 @@
+@tailwind base;
+@tailwind components;
+@tailwind utilities;
+
+/* Light mode (default) */
+:root {
+  --color-bg-dark: #f1f5f9;
+  --color-bg-card: #ffffff;
+  --color-bg-card-hover: #e2e8f0;
+  --color-text-primary: #0f172a;
+  --color-text-secondary: #475569;
+  --color-border: #cbd5e1;
+}
+
+/* Dark mode */
+.dark {
+  --color-bg-dark: #0f172a;
+  --color-bg-card: #1e293b;
+  --color-bg-card-hover: #334155;
+  --color-text-primary: #f8fafc;
+  --color-text-secondary: #94a3b8;
+  --color-border: #334155;
+}
+
+body {
+  font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, 'Helvetica Neue', Arial, sans-serif;
+  -webkit-font-smoothing: antialiased;
+  -moz-osx-font-smoothing: grayscale;
+}
+
+/* Custom scrollbar */
+::-webkit-scrollbar {
+  width: 8px;
+  height: 8px;
+}
+
+::-webkit-scrollbar-track {
+  background: var(--color-bg-card);
+}
+
+::-webkit-scrollbar-thumb {
+  background: #6366f1;
+  border-radius: 4px;
+}
+
+::-webkit-scrollbar-thumb:hover {
+  background: #4f46e5;
+}
+
+/* Selection */
+::selection {
+  background: rgba(99, 102, 241, 0.3);
+  color: var(--color-text-primary);
+}

frontend/src/main.tsx

@@ -0,0 +1,10 @@
+import { StrictMode } from 'react';
+import { createRoot } from 'react-dom/client';
+import App from './App';
+import './index.css';
+
+createRoot(document.getElementById('root')!).render(
+  <StrictMode>
+    <App />
+  </StrictMode>
+);

frontend/src/pages/About.tsx

@@ -0,0 +1,171 @@
+import { useQuery } from '@tanstack/react-query';
+import axios from 'axios';
+import { Search, FileText, Bot, Zap, Globe, BarChart3, CheckCircle, AlertTriangle, XCircle } from 'lucide-react';
+
+const API_BASE_URL = import.meta.env.VITE_API_URL || '/api';
+
+export function About() {
+  const { data: health } = useQuery({
+    queryKey: ['health'],
+    queryFn: async () => {
+      const response = await axios.get(`${API_BASE_URL}/health`);
+      return response.data;
+    },
+    refetchInterval: 30000,
+  });
+
+  const features = [
+    {
+      icon: Search,
+      title: 'Patent Retrieval',
+      description: 'Automated collection via SerpAPI\'s Google Patents',
+    },
+    {
+      icon: FileText,
+      title: 'Intelligent Parsing',
+      description: 'Extracts key sections from patent documents',
+    },
+    {
+      icon: Bot,
+      title: 'AI Analysis',
+      description: 'Deep analysis powered by Claude 3.5 Sonnet',
+    },
+    {
+      icon: Zap,
+      title: 'Batch Processing',
+      description: 'Analyze multiple companies concurrently',
+    },
+    {
+      icon: Globe,
+      title: 'REST API',
+      description: 'FastAPI web service for seamless integration',
+    },
+    {
+      icon: BarChart3,
+      title: 'Analytics',
+      description: 'Track and visualize historical analysis data',
+    },
+  ];
+
+  const techStack = [
+    { label: 'Backend', value: 'Python, FastAPI' },
+    { label: 'AI Model', value: 'Claude 3.5 Sonnet' },
+    { label: 'Database', value: 'PostgreSQL' },
+    { label: 'Frontend', value: 'React, TailwindCSS' },
+    { label: 'Data Source', value: 'SerpAPI Patents' },
+  ];
+
+  return (
+    <div className="space-y-8">
+      {/* Header */}
+      <div>
+        <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+          About SPARC
+        </h2>
+      </div>
+
+      <div className="grid grid-cols-1 lg:grid-cols-3 gap-8">
+        {/* Main Content */}
+        <div className="lg:col-span-2 space-y-6">
+          {/* Description */}
+          <p className="text-text-secondary leading-relaxed">
+            <strong className="text-text-primary">SPARC</strong> (Semiconductor Patent & Analytics Report Core)
+            is an AI-powered patent analysis platform that evaluates company performance by analyzing their
+            patent portfolios with cutting-edge language models.
+          </p>
+
+          {/* Features */}
+          <div>
+            <h3 className="text-lg font-semibold text-text-primary mb-4">Key Features</h3>
+            <div className="space-y-3">
+              {features.map(({ icon: Icon, title, description }) => (
+                <div
+                  key={title}
+                  className="flex items-start gap-4 py-3 border-b border-primary/10 last:border-0"
+                >
+                  <div className="flex-shrink-0">
+                    <Icon className="text-primary" size={20} />
+                  </div>
+                  <div>
+                    <div className="font-medium text-text-primary">{title}</div>
+                    <div className="text-sm text-text-secondary">{description}</div>
+                  </div>
+                </div>
+              ))}
+            </div>
+          </div>
+        </div>
+
+        {/* Sidebar */}
+        <div className="space-y-6">
+          {/* Tech Stack */}
+          <div className="bg-gradient-to-br from-primary/10 to-secondary/5 border border-primary/20 rounded-xl p-5">
+            <h3 className="font-semibold text-text-primary mb-4">Technology Stack</h3>
+            <div className="space-y-3">
+              {techStack.map(({ label, value }) => (
+                <div key={label}>
+                  <div className="text-primary text-sm">{label}</div>
+                  <div className="text-text-secondary text-sm">{value}</div>
+                </div>
+              ))}
+            </div>
+          </div>
+
+          {/* API Endpoints */}
+          <div className="bg-bg-card/60 border border-primary/15 rounded-xl p-5">
+            <h3 className="font-semibold text-text-primary mb-4">API Endpoints</h3>
+            <div className="space-y-2">
+              <code className="block bg-bg-dark px-3 py-2 rounded text-sm text-text-secondary">
+                http://localhost:8000/docs
+              </code>
+              <code className="block bg-bg-dark px-3 py-2 rounded text-sm text-text-secondary">
+                http://localhost:8000/health
+              </code>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      {/* System Status */}
+      <div>
+        <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
+          System Status
+        </h3>
+        <div className="grid grid-cols-1 md:grid-cols-3 gap-4">
+          <StatusCard
+            label="API"
+            status={health ? 'online' : 'offline'}
+          />
+          <StatusCard
+            label="Database"
+            status="configured"
+          />
+          <StatusCard
+            label="Dashboard"
+            status="online"
+          />
+        </div>
+      </div>
+    </div>
+  );
+}
+
+function StatusCard({ label, status }: { label: string; status: 'online' | 'offline' | 'configured' }) {
+  const statusConfig = {
+    online: { icon: CheckCircle, color: 'text-success', bg: 'bg-success' },
+    offline: { icon: XCircle, color: 'text-error', bg: 'bg-error' },
+    configured: { icon: AlertTriangle, color: 'text-warning', bg: 'bg-warning' },
+  };
+
+  const { icon: Icon, color, bg } = statusConfig[status];
+
+  return (
+    <div className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-5 text-center">
+      <div className={`inline-flex items-center justify-center w-8 h-8 rounded-full ${bg}/20 mb-2`}>
+        <Icon className={color} size={20} />
+      </div>
+      <div className="text-sm text-text-secondary uppercase tracking-wide">{label}</div>
+      <div className={`font-semibold ${color} capitalize`}>{status}</div>
+    </div>
+  );
+}

frontend/src/pages/AdminUsers.tsx

@@ -0,0 +1,183 @@
+import { useState } from 'react';
+import { useQuery, useMutation, useQueryClient } from '@tanstack/react-query';
+import { adminApi } from '../api/client';
+import { useAuth } from '../context/AuthContext';
+import { Users, Shield, User, Trash2, AlertCircle } from 'lucide-react';
+import type { User as UserType } from '../types';
+
+export function AdminUsers() {
+  const { user: currentUser } = useAuth();
+  const queryClient = useQueryClient();
+  const [deleteConfirm, setDeleteConfirm] = useState<number | null>(null);
+
+  const { data: users, isLoading, isError } = useQuery({
+    queryKey: ['admin-users'],
+    queryFn: () => adminApi.listUsers(),
+  });
+
+  const updateRoleMutation = useMutation({
+    mutationFn: ({ userId, role }: { userId: number; role: 'admin' | 'user' }) =>
+      adminApi.updateUserRole(userId, role),
+    onSuccess: () => {
+      queryClient.invalidateQueries({ queryKey: ['admin-users'] });
+    },
+  });
+
+  const deleteMutation = useMutation({
+    mutationFn: (userId: number) => adminApi.deleteUser(userId),
+    onSuccess: () => {
+      queryClient.invalidateQueries({ queryKey: ['admin-users'] });
+      setDeleteConfirm(null);
+    },
+  });
+
+  const handleRoleChange = (user: UserType) => {
+    const newRole = user.role === 'admin' ? 'user' : 'admin';
+    updateRoleMutation.mutate({ userId: user.id, role: newRole });
+  };
+
+  const handleDelete = (userId: number) => {
+    deleteMutation.mutate(userId);
+  };
+
+  if (isLoading) {
+    return (
+      <div className="flex items-center justify-center min-h-[400px]">
+        <div className="animate-spin rounded-full h-12 w-12 border-t-2 border-b-2 border-primary"></div>
+      </div>
+    );
+  }
+
+  if (isError) {
+    return (
+      <div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-xl px-4 py-3">
+        <AlertCircle size={18} />
+        <span>Failed to load users.</span>
+      </div>
+    );
+  }
+
+  return (
+    <div className="space-y-6">
+      {/* Header */}
+      <div className="flex items-center justify-between">
+        <div>
+          <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+            User Management
+          </h2>
+          <p className="text-text-secondary">Manage user accounts and permissions.</p>
+        </div>
+        <div className="flex items-center gap-2 bg-primary/10 border border-primary/20 rounded-xl px-4 py-2">
+          <Users size={18} className="text-primary" />
+          <span className="text-text-primary font-semibold">{users?.length || 0} Users</span>
+        </div>
+      </div>
+
+      {/* Users Table */}
+      <div className="bg-bg-card/60 border border-primary/15 rounded-2xl overflow-hidden">
+        <div className="overflow-x-auto">
+          <table className="w-full">
+            <thead>
+              <tr className="border-b border-primary/10">
+                <th className="text-left px-6 py-4 text-sm font-semibold text-text-secondary uppercase tracking-wider">
+                  User
+                </th>
+                <th className="text-left px-6 py-4 text-sm font-semibold text-text-secondary uppercase tracking-wider">
+                  Role
+                </th>
+                <th className="text-left px-6 py-4 text-sm font-semibold text-text-secondary uppercase tracking-wider">
+                  Created
+                </th>
+                <th className="text-right px-6 py-4 text-sm font-semibold text-text-secondary uppercase tracking-wider">
+                  Actions
+                </th>
+              </tr>
+            </thead>
+            <tbody className="divide-y divide-primary/10">
+              {users?.map((user) => (
+                <tr key={user.id} className="hover:bg-bg-card-hover/50 transition-colors">
+                  <td className="px-6 py-4">
+                    <div className="flex items-center gap-3">
+                      <div className="w-10 h-10 rounded-full bg-gradient-to-br from-primary/20 to-secondary/20 flex items-center justify-center">
+                        {user.role === 'admin' ? (
+                          <Shield className="text-primary" size={18} />
+                        ) : (
+                          <User className="text-secondary" size={18} />
+                        )}
+                      </div>
+                      <div>
+                        <div className="font-medium text-text-primary">{user.email}</div>
+                        {user.id === currentUser?.id && (
+                          <span className="text-xs text-primary">(You)</span>
+                        )}
+                      </div>
+                    </div>
+                  </td>
+                  <td className="px-6 py-4">
+                    <span
+                      className={`inline-flex items-center gap-1 px-3 py-1 rounded-full text-xs font-semibold uppercase ${
+                        user.role === 'admin'
+                          ? 'bg-primary/20 text-primary border border-primary/30'
+                          : 'bg-secondary/20 text-secondary border border-secondary/30'
+                      }`}
+                    >
+                      {user.role === 'admin' ? <Shield size={12} /> : <User size={12} />}
+                      {user.role}
+                    </span>
+                  </td>
+                  <td className="px-6 py-4 text-text-secondary">
+                    {new Date(user.created_at).toLocaleDateString()}
+                  </td>
+                  <td className="px-6 py-4">
+                    <div className="flex items-center justify-end gap-2">
+                      {user.id !== currentUser?.id && (
+                        <>
+                          <button
+                            onClick={() => handleRoleChange(user)}
+                            disabled={updateRoleMutation.isPending}
+                            className={`px-3 py-1.5 rounded-lg text-sm font-medium transition-all ${
+                              user.role === 'admin'
+                                ? 'bg-secondary/10 text-secondary hover:bg-secondary/20 border border-secondary/30'
+                                : 'bg-primary/10 text-primary hover:bg-primary/20 border border-primary/30'
+                            } disabled:opacity-50`}
+                          >
+                            {user.role === 'admin' ? 'Demote' : 'Promote'}
+                          </button>
+
+                          {deleteConfirm === user.id ? (
+                            <div className="flex items-center gap-1">
+                              <button
+                                onClick={() => handleDelete(user.id)}
+                                disabled={deleteMutation.isPending}
+                                className="px-3 py-1.5 rounded-lg text-sm font-medium bg-error text-white hover:bg-error/80 transition-all disabled:opacity-50"
+                              >
+                                Confirm
+                              </button>
+                              <button
+                                onClick={() => setDeleteConfirm(null)}
+                                className="px-3 py-1.5 rounded-lg text-sm font-medium bg-bg-card-hover text-text-secondary hover:text-text-primary transition-all"
+                              >
+                                Cancel
+                              </button>
+                            </div>
+                          ) : (
+                            <button
+                              onClick={() => setDeleteConfirm(user.id)}
+                              className="p-1.5 rounded-lg text-error/70 hover:text-error hover:bg-error/10 transition-all"
+                            >
+                              <Trash2 size={18} />
+                            </button>
+                          )}
+                        </>
+                      )}
+                    </div>
+                  </td>
+                </tr>
+              ))}
+            </tbody>
+          </table>
+        </div>
+      </div>
+    </div>
+  );
+}

frontend/src/pages/Analysis.tsx

@@ -0,0 +1,185 @@
+import { useState } from 'react';
+import { useMutation, useQuery } from '@tanstack/react-query';
+import { analysisApi, exportApi } from '../api/client';
+import { Search, CheckCircle, AlertCircle, Clock, FileText, Download, ChevronDown } from 'lucide-react';
+import type { CompanyAnalysis } from '../types';
+
+export function Analysis() {
+  const [companyName, setCompanyName] = useState('');
+  const [selectedModel, setSelectedModel] = useState('');
+  const [result, setResult] = useState<CompanyAnalysis | null>(null);
+
+  const modelsQuery = useQuery({
+    queryKey: ['models'],
+    queryFn: () => analysisApi.listModels(),
+  });
+
+  const mutation = useMutation({
+    mutationFn: (name: string) => analysisApi.analyzeCompany(name, selectedModel || undefined),
+    onSuccess: (data) => setResult(data),
+  });
+
+  const handleSubmit = (e: React.FormEvent) => {
+    e.preventDefault();
+    if (companyName.trim()) {
+      mutation.mutate(companyName.trim());
+    }
+  };
+
+  return (
+    <div className="space-y-6">
+      {/* Header */}
+      <div>
+        <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+          Single Company Analysis
+        </h2>
+        <p className="text-text-secondary">
+          Analyze a company's patent portfolio using AI-powered insights.
+        </p>
+      </div>
+
+      {/* Search Form */}
+      <form onSubmit={handleSubmit} className="space-y-4">
+        <div className="flex gap-4">
+          <div className="flex-1 relative">
+            <Search className="absolute left-4 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+            <input
+              type="text"
+              value={companyName}
+              onChange={(e) => setCompanyName(e.target.value)}
+              placeholder="Enter company name (e.g., nvidia, intel, amd)"
+              className="w-full bg-bg-card/80 border border-primary/30 rounded-xl pl-12 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+            />
+          </div>
+          <button
+            type="submit"
+            disabled={mutation.isPending || !companyName.trim()}
+            className="bg-gradient-to-r from-primary to-primary-dark text-white font-semibold py-3 px-6 rounded-xl hover:shadow-lg hover:shadow-primary/30 transition-all disabled:opacity-50 disabled:cursor-not-allowed flex items-center gap-2"
+          >
+            {mutation.isPending ? (
+              <div className="animate-spin rounded-full h-5 w-5 border-t-2 border-b-2 border-white"></div>
+            ) : (
+              <>
+                <Search size={18} />
+                Analyze
+              </>
+            )}
+          </button>
+        </div>
+
+        {/* Model Selector */}
+        <div className="flex items-center gap-3">
+          <label className="text-sm font-medium text-text-secondary whitespace-nowrap">
+            LLM Model
+          </label>
+          <div className="relative flex-1 max-w-xs">
+            <select
+              value={selectedModel}
+              onChange={(e) => setSelectedModel(e.target.value)}
+              className="w-full appearance-none bg-bg-card/80 border border-primary/30 rounded-lg pl-3 pr-8 py-2 text-sm text-text-primary focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all cursor-pointer"
+            >
+              <option value="">
+                {modelsQuery.data ? `Default (${modelsQuery.data.default})` : 'Default'}
+              </option>
+              {modelsQuery.data?.models.map((m) => (
+                <option key={m.id} value={m.id}>
+                  {m.name} ({m.provider})
+                </option>
+              ))}
+            </select>
+            <ChevronDown className="absolute right-2 top-1/2 -translate-y-1/2 text-text-secondary pointer-events-none" size={16} />
+          </div>
+        </div>
+      </form>
+
+      {/* Error */}
+      {mutation.isError && (
+        <div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-xl px-4 py-3">
+          <AlertCircle size={18} />
+          <span>Analysis failed. Please try again.</span>
+        </div>
+      )}
+
+      {/* Results */}
+      {result && (
+        <div className="space-y-6">
+          {/* Success/Failure Status */}
+          {result.success ? (
+            <div className="flex items-center gap-2 bg-success/10 border border-success/20 text-success rounded-xl px-4 py-3">
+              <CheckCircle size={18} />
+              <span>Analysis complete for {result.company_name.toUpperCase()}</span>
+            </div>
+          ) : (
+            <div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-xl px-4 py-3">
+              <AlertCircle size={18} />
+              <span>Analysis failed: {result.error}</span>
+            </div>
+          )}
+
+          {/* Metrics */}
+          <div className="grid grid-cols-1 md:grid-cols-3 gap-4">
+            <MetricCard
+              icon={FileText}
+              label="Patents Found"
+              value={result.patent_count.toString()}
+            />
+            <MetricCard
+              icon={CheckCircle}
+              label="Analysis Status"
+              value={result.success ? 'Complete' : 'Failed'}
+            />
+            <MetricCard
+              icon={Clock}
+              label="Timestamp"
+              value={new Date(result.timestamp).toLocaleTimeString()}
+            />
+          </div>
+
+          {/* Analysis Content */}
+          {result.success && result.analysis && (
+            <div className="bg-bg-card/60 backdrop-blur-lg border border-primary/15 rounded-2xl p-6">
+              <div className="flex items-center justify-between border-b-2 border-primary/30 pb-2 mb-4">
+                <h3 className="text-lg font-semibold text-text-primary">
+                  AI Analysis Results
+                </h3>
+                <div className="flex items-center gap-2">
+                  <button
+                    onClick={() => exportApi.exportCsv(result.company_name)}
+                    className="flex items-center gap-2 text-sm bg-primary/20 hover:bg-primary/30 text-primary font-medium px-3 py-1.5 rounded-lg transition-colors"
+                  >
+                    <Download size={14} />
+                    Export CSV
+                  </button>
+                  <button
+                    onClick={() => exportApi.exportPdf(result.company_name)}
+                    className="flex items-center gap-2 text-sm bg-primary/20 hover:bg-primary/30 text-primary font-medium px-3 py-1.5 rounded-lg transition-colors"
+                  >
+                    <FileText size={14} />
+                    Export PDF
+                  </button>
+                </div>
+              </div>
+              <div className="prose prose-invert max-w-none">
+                <div className="text-text-primary whitespace-pre-wrap leading-relaxed">
+                  {result.analysis}
+                </div>
+              </div>
+            </div>
+          )}
+        </div>
+      )}
+    </div>
+  );
+}
+
+function MetricCard({ icon: Icon, label, value }: { icon: typeof FileText; label: string; value: string }) {
+  return (
+    <div className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-5 text-center">
+      <Icon className="mx-auto mb-2 text-primary" size={24} />
+      <div className="text-2xl font-bold bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent">
+        {value}
+      </div>
+      <div className="text-sm text-text-secondary uppercase tracking-wide mt-1">{label}</div>
+    </div>
+  );
+}

frontend/src/pages/Analytics.tsx

@@ -0,0 +1,304 @@
+import { useState } from 'react';
+import { useQuery } from '@tanstack/react-query';
+import { analyticsApi } from '../api/client';
+import { AlertCircle, Database } from 'lucide-react';
+import { PieChart, Pie, Cell, BarChart, Bar, LineChart, Line, XAxis, YAxis, Tooltip, ResponsiveContainer, Legend } from 'recharts';
+import { useChartTheme } from '../context/useChartTheme';
+
+const COLORS = ['#6366f1', '#0ea5e9', '#10b981', '#f59e0b', '#ef4444', '#8b5cf6', '#ec4899', '#14b8a6'];
+
+export function AnalyticsPage() {
+  const [days, setDays] = useState(30);
+  const chartTheme = useChartTheme();
+
+  const { data, isLoading, isError, refetch } = useQuery({
+    queryKey: ['analytics', days],
+    queryFn: () => analyticsApi.getAnalytics(days),
+  });
+
+  const trendsQuery = useQuery({
+    queryKey: ['analytics-trends', days],
+    queryFn: () => analyticsApi.getTrends(days),
+  });
+
+  if (isLoading) {
+    return (
+      <div className="space-y-6">
+        <div>
+          <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+            Analytics Dashboard
+          </h2>
+          <p className="text-text-secondary">Loading analytics data...</p>
+        </div>
+        {/* Skeleton cards */}
+        <div className="grid grid-cols-1 md:grid-cols-3 gap-4">
+          {[1, 2, 3].map((i) => (
+            <div key={i} className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-5 text-center animate-pulse">
+              <div className="h-9 w-16 bg-primary/20 rounded mx-auto mb-2" />
+              <div className="h-4 w-24 bg-primary/10 rounded mx-auto" />
+            </div>
+          ))}
+        </div>
+        {/* Skeleton charts */}
+        <div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
+          {[1, 2].map((i) => (
+            <div key={i} className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6 animate-pulse">
+              <div className="h-5 w-40 bg-primary/20 rounded mb-4" />
+              <div className="h-[300px] bg-primary/5 rounded" />
+            </div>
+          ))}
+        </div>
+      </div>
+    );
+  }
+
+  if (isError) {
+    return (
+      <div className="space-y-6">
+        <div>
+          <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+            Analytics Dashboard
+          </h2>
+        </div>
+        <div className="bg-gradient-to-br from-primary/10 to-secondary/5 border border-primary/20 rounded-xl p-6">
+          <div className="flex items-center gap-3 text-warning mb-2">
+            <Database size={24} />
+            <span className="font-semibold">Unable to Load Analytics</span>
+          </div>
+          <p className="text-text-secondary">
+            Could not connect to the analytics database. Ensure PostgreSQL is running and
+            <code className="bg-bg-card px-2 py-1 rounded mx-1">DATABASE_URL</code> is configured correctly.
+          </p>
+          <button
+            onClick={() => refetch()}
+            className="mt-3 text-sm bg-primary/20 hover:bg-primary/30 text-primary font-medium px-4 py-2 rounded-lg transition-colors"
+          >
+            Retry
+          </button>
+        </div>
+      </div>
+    );
+  }
+
+  if (!data || (data.total_messages === 0 && data.by_company.length === 0)) {
+    return (
+      <div className="space-y-6">
+        <div>
+          <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+            Analytics Dashboard
+          </h2>
+          <p className="text-text-secondary">Track historical analysis data and view insights.</p>
+        </div>
+        <div className="flex items-center gap-2 bg-secondary/10 border border-secondary/20 text-secondary rounded-xl px-4 py-3">
+          <AlertCircle size={18} />
+          <span>No analytics data available yet. Run some analyses first!</span>
+        </div>
+      </div>
+    );
+  }
+
+  const companyData = data.by_company.map((c) => ({
+    name: (c.company_name || 'Unknown').toUpperCase(),
+    value: c.count,
+  }));
+
+  const typeData = data.by_type.map((t) => ({
+    name: t.analysis_type || 'Unknown',
+    count: t.count,
+  }));
+
+  return (
+    <div className="space-y-6">
+      {/* Header */}
+      <div className="flex flex-wrap items-center justify-between gap-4">
+        <div>
+          <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+            Analytics Dashboard
+          </h2>
+          <p className="text-text-secondary">Track historical analysis data and view insights.</p>
+        </div>
+
+        {/* Time Range Selector */}
+        <select
+          value={days}
+          onChange={(e) => setDays(Number(e.target.value))}
+          className="bg-bg-card/80 border border-primary/30 rounded-xl px-4 py-2 text-text-primary focus:outline-none focus:border-primary"
+        >
+          <option value={7}>Last 7 days</option>
+          <option value={14}>Last 14 days</option>
+          <option value={30}>Last 30 days</option>
+          <option value={90}>Last 90 days</option>
+        </select>
+      </div>
+
+      {/* Summary Metrics */}
+      <div className="grid grid-cols-1 md:grid-cols-3 gap-4">
+        <MetricCard label="Total Analyses" value={data.total_messages} />
+        <MetricCard label="Companies Analyzed" value={data.by_company.length} />
+        <MetricCard label="Analysis Types" value={data.by_type.length} />
+      </div>
+
+      {/* Charts */}
+      <div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
+        {/* Pie Chart - Distribution by Company */}
+        {companyData.length > 0 && (
+          <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6">
+            <h3 className="text-lg font-semibold text-text-primary mb-4">Distribution by Company</h3>
+            <ResponsiveContainer width="100%" height={300}>
+              <PieChart>
+                <Pie
+                  data={companyData}
+                  cx="50%"
+                  cy="50%"
+                  innerRadius={60}
+                  outerRadius={100}
+                  paddingAngle={2}
+                  dataKey="value"
+                  label={({ name, percent }) => `${name} ${(percent * 100).toFixed(0)}%`}
+                  labelLine={false}
+                >
+                  {companyData.map((_, index) => (
+                    <Cell key={`cell-${index}`} fill={COLORS[index % COLORS.length]} />
+                  ))}
+                </Pie>
+                <Tooltip
+                  contentStyle={chartTheme.tooltipContentStyle}
+                />
+                <Legend />
+              </PieChart>
+            </ResponsiveContainer>
+          </div>
+        )}
+
+        {/* Bar Chart - Analysis Types */}
+        {typeData.length > 0 && (
+          <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6">
+            <h3 className="text-lg font-semibold text-text-primary mb-4">Analysis Types</h3>
+            <ResponsiveContainer width="100%" height={300}>
+              <BarChart data={typeData}>
+                <XAxis dataKey="name" stroke={chartTheme.axisStroke} fontSize={12} />
+                <YAxis stroke={chartTheme.axisStroke} fontSize={12} />
+                <Tooltip
+                  contentStyle={chartTheme.tooltipContentStyle}
+                  labelStyle={chartTheme.tooltipLabelStyle}
+                />
+                <Bar dataKey="count" fill="#6366f1" radius={[4, 4, 0, 0]} />
+              </BarChart>
+            </ResponsiveContainer>
+          </div>
+        )}
+      </div>
+
+      {/* Trend Charts */}
+      {trendsQuery.data && (
+        <div className="space-y-6">
+          <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2">
+            Trends Over Time
+          </h3>
+
+          <div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
+            {/* Patent count over time per company (line chart) */}
+            {trendsQuery.data.by_month.length > 0 && (() => {
+              // Pivot data: each month as a row, companies as columns
+              const companies = [...new Set(trendsQuery.data!.by_month.map(d => d.company_name))];
+              const months = [...new Set(trendsQuery.data!.by_month.map(d => d.month))].sort();
+              const pivoted = months.map(month => {
+                const row: Record<string, string | number> = { month };
+                for (const c of companies) {
+                  const entry = trendsQuery.data!.by_month.find(d => d.month === month && d.company_name === c);
+                  row[c] = entry?.count || 0;
+                }
+                return row;
+              });
+
+              return (
+                <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6">
+                  <h4 className="text-md font-semibold text-text-primary mb-4">Analyses per Company Over Time</h4>
+                  <ResponsiveContainer width="100%" height={300}>
+                    <LineChart data={pivoted}>
+                      <XAxis dataKey="month" stroke={chartTheme.axisStroke} fontSize={12} />
+                      <YAxis stroke={chartTheme.axisStroke} fontSize={12} />
+                      <Tooltip
+                        contentStyle={chartTheme.tooltipContentStyle}
+                        labelStyle={chartTheme.tooltipLabelStyle}
+                      />
+                      <Legend />
+                      {companies.map((company, idx) => (
+                        <Line
+                          key={company}
+                          type="monotone"
+                          dataKey={company}
+                          stroke={COLORS[idx % COLORS.length]}
+                          strokeWidth={2}
+                          dot={{ r: 4 }}
+                          name={company.toUpperCase()}
+                        />
+                      ))}
+                    </LineChart>
+                  </ResponsiveContainer>
+                </div>
+              );
+            })()}
+
+            {/* Analysis type distribution over time (stacked bar) */}
+            {trendsQuery.data.by_type_over_time.length > 0 && (() => {
+              const types = [...new Set(trendsQuery.data!.by_type_over_time.map(d => d.analysis_type))];
+              const months = [...new Set(trendsQuery.data!.by_type_over_time.map(d => d.month))].sort();
+              const pivoted = months.map(month => {
+                const row: Record<string, string | number> = { month };
+                for (const t of types) {
+                  const entry = trendsQuery.data!.by_type_over_time.find(d => d.month === month && d.analysis_type === t);
+                  row[t] = entry?.count || 0;
+                }
+                return row;
+              });
+
+              return (
+                <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6">
+                  <h4 className="text-md font-semibold text-text-primary mb-4">Analysis Types Over Time</h4>
+                  <ResponsiveContainer width="100%" height={300}>
+                    <BarChart data={pivoted}>
+                      <XAxis dataKey="month" stroke={chartTheme.axisStroke} fontSize={12} />
+                      <YAxis stroke={chartTheme.axisStroke} fontSize={12} />
+                      <Tooltip
+                        contentStyle={chartTheme.tooltipContentStyle}
+                        labelStyle={chartTheme.tooltipLabelStyle}
+                      />
+                      <Legend />
+                      {types.map((type, idx) => (
+                        <Bar
+                          key={type}
+                          dataKey={type}
+                          stackId="types"
+                          fill={COLORS[idx % COLORS.length]}
+                          name={type}
+                        />
+                      ))}
+                    </BarChart>
+                  </ResponsiveContainer>
+                </div>
+              );
+            })()}
+          </div>
+
+          {trendsQuery.data.by_month.length === 0 && (
+            <div className="text-text-secondary text-center py-8">
+              No trend data available yet. Run analyses over multiple days to see trends.
+            </div>
+          )}
+        </div>
+      )}
+    </div>
+  );
+}
+
+function MetricCard({ label, value }: { label: string; value: number }) {
+  return (
+    <div className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-5 text-center">
+      <div className="text-3xl font-bold bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent">
+        {value}
+      </div>
+      <div className="text-sm text-text-secondary uppercase tracking-wide mt-1">{label}</div>
+    </div>
+  );
+}

frontend/src/pages/Batch.tsx

@@ -0,0 +1,430 @@
+import { useState } from 'react';
+import { useMutation, useQuery } from '@tanstack/react-query';
+import { analysisApi } from '../api/client';
+import { Rocket, CheckCircle, AlertCircle, ChevronDown, ChevronUp, RefreshCw, Inbox } from 'lucide-react';
+import { BarChart, Bar, XAxis, YAxis, Tooltip, ResponsiveContainer, Cell } from 'recharts';
+import { useChartTheme } from '../context/useChartTheme';
+import type { BatchAnalysisResult } from '../types';
+
+export function Batch() {
+  const [companiesInput, setCompaniesInput] = useState('');
+  const [maxWorkers, setMaxWorkers] = useState(3);
+  const [selectedModel, setSelectedModel] = useState('');
+  const [result, setResult] = useState<BatchAnalysisResult | null>(null);
+  const [expandedItems, setExpandedItems] = useState<Set<string>>(new Set());
+
+  const chartTheme = useChartTheme();
+
+  const modelsQuery = useQuery({
+    queryKey: ['models'],
+    queryFn: () => analysisApi.listModels(),
+  });
+
+  const jobsQuery = useQuery({
+    queryKey: ['jobs'],
+    queryFn: () => analysisApi.listJobs(undefined, 20),
+  });
+
+  const mutation = useMutation({
+    mutationFn: ({ companies, workers }: { companies: string[]; workers: number }) =>
+      analysisApi.analyzeBatch(companies, workers, selectedModel || undefined),
+    onSuccess: (data) => {
+      setResult(data);
+      jobsQuery.refetch();
+    },
+  });
+
+  const handleSubmit = (e: React.FormEvent) => {
+    e.preventDefault();
+    const companies = companiesInput
+      .split(/[,\n]/)
+      .map((c) => c.trim())
+      .filter((c) => c.length > 0);
+
+    if (companies.length > 0) {
+      mutation.mutate({ companies, workers: maxWorkers });
+    }
+  };
+
+  const toggleExpand = (company: string) => {
+    const newExpanded = new Set(expandedItems);
+    if (newExpanded.has(company)) {
+      newExpanded.delete(company);
+    } else {
+      newExpanded.add(company);
+    }
+    setExpandedItems(newExpanded);
+  };
+
+  const chartData = result?.results.map((r) => ({
+    name: r.company_name.toUpperCase(),
+    patents: r.patent_count,
+    success: r.success,
+  }));
+
+  return (
+    <div className="space-y-6">
+      {/* Header */}
+      <div>
+        <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+          Batch Company Analysis
+        </h2>
+        <p className="text-text-secondary">
+          Analyze multiple companies simultaneously for comparative insights.
+        </p>
+      </div>
+
+      {/* Input Form */}
+      <form onSubmit={handleSubmit} className="grid grid-cols-1 md:grid-cols-3 gap-4">
+        <div className="md:col-span-2">
+          <textarea
+            value={companiesInput}
+            onChange={(e) => setCompaniesInput(e.target.value)}
+            placeholder="Enter company names (one per line or comma-separated):&#10;nvidia&#10;amd&#10;intel&#10;qualcomm"
+            rows={6}
+            className="w-full bg-bg-card/80 border border-primary/30 rounded-xl px-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all resize-none"
+          />
+        </div>
+
+        <div className="space-y-4">
+          <div>
+            <label className="block text-sm font-medium text-text-secondary mb-2">
+              Concurrent Workers
+            </label>
+            <input
+              type="range"
+              min={1}
+              max={5}
+              value={maxWorkers}
+              onChange={(e) => setMaxWorkers(Number(e.target.value))}
+              className="w-full accent-primary"
+            />
+            <div className="text-center text-text-primary font-semibold">{maxWorkers}</div>
+          </div>
+
+          <div>
+            <label className="block text-sm font-medium text-text-secondary mb-2">
+              LLM Model
+            </label>
+            <div className="relative">
+              <select
+                value={selectedModel}
+                onChange={(e) => setSelectedModel(e.target.value)}
+                className="w-full appearance-none bg-bg-card/80 border border-primary/30 rounded-lg pl-3 pr-8 py-2 text-sm text-text-primary focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all cursor-pointer"
+              >
+                <option value="">
+                  {modelsQuery.data ? `Default (${modelsQuery.data.default})` : 'Default'}
+                </option>
+                {modelsQuery.data?.models.map((m) => (
+                  <option key={m.id} value={m.id}>
+                    {m.name} ({m.provider})
+                  </option>
+                ))}
+              </select>
+              <ChevronDown className="absolute right-2 top-1/2 -translate-y-1/2 text-text-secondary pointer-events-none" size={16} />
+            </div>
+          </div>
+
+          <button
+            type="submit"
+            disabled={mutation.isPending || !companiesInput.trim()}
+            className="w-full bg-gradient-to-r from-primary to-primary-dark text-white font-semibold py-3 px-6 rounded-xl hover:shadow-lg hover:shadow-primary/30 transition-all disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
+          >
+            {mutation.isPending ? (
+              <div className="animate-spin rounded-full h-5 w-5 border-t-2 border-b-2 border-white"></div>
+            ) : (
+              <>
+                <Rocket size={18} />
+                Run Batch Analysis
+              </>
+            )}
+          </button>
+        </div>
+      </form>
+
+      {/* Progress */}
+      {mutation.isPending && (
+        <div className="bg-bg-card/60 border border-primary/15 rounded-xl p-4">
+          <div className="flex items-center gap-2 text-secondary">
+            <div className="animate-spin rounded-full h-4 w-4 border-t-2 border-b-2 border-secondary"></div>
+            <span>Analyzing companies...</span>
+          </div>
+        </div>
+      )}
+
+      {/* Error */}
+      {mutation.isError && (
+        <div className="bg-error/10 border border-error/20 rounded-xl px-4 py-3">
+          <div className="flex items-center gap-2 text-error">
+            <AlertCircle size={18} />
+            <span className="font-semibold">Batch analysis failed</span>
+          </div>
+          <p className="text-text-secondary text-sm mt-1 ml-7">
+            {mutation.error instanceof Error ? mutation.error.message : 'An unexpected error occurred.'}
+            {' '}Check your connection and try again.
+          </p>
+          <div className="ml-7 mt-2 flex items-center gap-3">
+            <button
+              onClick={() => {
+                const companies = companiesInput
+                  .split(/[,\n]/)
+                  .map((c) => c.trim())
+                  .filter((c) => c.length > 0);
+                if (companies.length > 0) {
+                  mutation.mutate({ companies, workers: maxWorkers });
+                }
+              }}
+              className="text-sm text-primary hover:text-primary-dark underline flex items-center gap-1"
+            >
+              <RefreshCw size={14} />
+              Retry
+            </button>
+            <button
+              onClick={() => mutation.reset()}
+              className="text-sm text-text-secondary hover:text-text-primary underline"
+            >
+              Dismiss
+            </button>
+          </div>
+        </div>
+      )}
+
+      {/* Results */}
+      {result && (
+        <div className="space-y-6">
+          {/* Summary Metrics */}
+          <div>
+            <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
+              Results Summary
+            </h3>
+            <div className="grid grid-cols-2 md:grid-cols-4 gap-4">
+              <SummaryCard label="Total Companies" value={result.total_companies} />
+              <SummaryCard label="Successful" value={result.successful} color="success" />
+              <SummaryCard label="Failed" value={result.failed} color="error" />
+              <SummaryCard
+                label="Success Rate"
+                value={`${Math.round((result.successful / result.total_companies) * 100)}%`}
+              />
+            </div>
+          </div>
+
+          {/* Chart */}
+          {chartData && chartData.length > 0 && (
+            <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6">
+              <ResponsiveContainer width="100%" height={300}>
+                <BarChart data={chartData}>
+                  <XAxis dataKey="name" stroke={chartTheme.axisStroke} fontSize={12} />
+                  <YAxis stroke={chartTheme.axisStroke} fontSize={12} />
+                  <Tooltip
+                    contentStyle={chartTheme.tooltipContentStyle}
+                    labelStyle={chartTheme.tooltipLabelStyle}
+                  />
+                  <Bar dataKey="patents" radius={[4, 4, 0, 0]}>
+                    {chartData.map((entry, index) => (
+                      <Cell
+                        key={`cell-${index}`}
+                        fill={entry.success ? '#10b981' : '#ef4444'}
+                      />
+                    ))}
+                  </Bar>
+                </BarChart>
+              </ResponsiveContainer>
+            </div>
+          )}
+
+          {/* Detailed Results */}
+          <div>
+            <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
+              Detailed Results
+            </h3>
+            <div className="space-y-3">
+              {result.results.map((r) => (
+                <div
+                  key={r.company_name}
+                  className="bg-bg-card/60 border border-primary/15 rounded-xl overflow-hidden"
+                >
+                  <button
+                    onClick={() => toggleExpand(r.company_name)}
+                    className="w-full flex items-center justify-between p-4 hover:bg-bg-card-hover transition-colors"
+                  >
+                    <div className="flex items-center gap-3">
+                      {r.success ? (
+                        <CheckCircle className="text-success" size={20} />
+                      ) : (
+                        <AlertCircle className="text-error" size={20} />
+                      )}
+                      <span className="font-semibold text-text-primary">
+                        {r.company_name.toUpperCase()}
+                      </span>
+                      <span className="text-text-secondary">
+                        {r.patent_count} patents
+                      </span>
+                    </div>
+                    {expandedItems.has(r.company_name) ? (
+                      <ChevronUp className="text-text-secondary" size={20} />
+                    ) : (
+                      <ChevronDown className="text-text-secondary" size={20} />
+                    )}
+                  </button>
+                  {expandedItems.has(r.company_name) && (
+                    <div className="border-t border-primary/10 p-4 bg-bg-dark/40">
+                      {r.success ? (
+                        <div className="text-text-primary whitespace-pre-wrap leading-relaxed">
+                          {r.analysis}
+                        </div>
+                      ) : (
+                        <div className="text-error">{r.error}</div>
+                      )}
+                    </div>
+                  )}
+                </div>
+              ))}
+            </div>
+          </div>
+        </div>
+      )}
+
+      {/* Job History */}
+      <div>
+        <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
+          Job History
+        </h3>
+
+        {/* Loading skeleton */}
+        {jobsQuery.isLoading && (
+          <div className="space-y-3">
+            {[...Array(3)].map((_, i) => (
+              <div
+                key={i}
+                className="bg-bg-card/60 border border-primary/15 rounded-xl p-4 animate-pulse"
+              >
+                <div className="flex items-center justify-between">
+                  <div className="flex items-center gap-3">
+                    <div className="h-5 w-5 rounded-full bg-primary/20" />
+                    <div className="h-4 w-32 rounded bg-primary/20" />
+                    <div className="h-4 w-20 rounded bg-primary/10" />
+                  </div>
+                  <div className="h-6 w-20 rounded-full bg-primary/15" />
+                </div>
+                <div className="mt-3 flex gap-4">
+                  <div className="h-3 w-24 rounded bg-primary/10" />
+                  <div className="h-3 w-16 rounded bg-primary/10" />
+                </div>
+              </div>
+            ))}
+          </div>
+        )}
+
+        {/* Job history error */}
+        {jobsQuery.isError && (
+          <div className="bg-error/10 border border-error/20 rounded-xl px-4 py-3">
+            <div className="flex items-center gap-2 text-error">
+              <AlertCircle size={18} />
+              <span className="font-semibold">Failed to load job history</span>
+            </div>
+            <p className="text-text-secondary text-sm mt-1 ml-7">
+              {jobsQuery.error instanceof Error ? jobsQuery.error.message : 'Could not retrieve past jobs.'}
+            </p>
+            <button
+              onClick={() => jobsQuery.refetch()}
+              className="ml-7 mt-2 text-sm text-primary hover:text-primary-dark underline flex items-center gap-1"
+            >
+              <RefreshCw size={14} />
+              Retry
+            </button>
+          </div>
+        )}
+
+        {/* Empty state */}
+        {jobsQuery.isSuccess && jobsQuery.data.length === 0 && !result && (
+          <div className="bg-bg-card/60 border border-primary/15 border-dashed rounded-xl p-8 text-center">
+            <Inbox className="mx-auto text-text-secondary/40 mb-3" size={40} />
+            <p className="text-text-secondary font-medium">No batch jobs yet</p>
+            <p className="text-text-secondary/70 text-sm mt-1">
+              Submit a batch analysis above to get started. Your job history will appear here.
+            </p>
+          </div>
+        )}
+
+        {/* Job list */}
+        {jobsQuery.isSuccess && jobsQuery.data.length > 0 && (
+          <div className="space-y-3">
+            {jobsQuery.data.map((job) => (
+              <div
+                key={job.job_id}
+                className="bg-bg-card/60 border border-primary/15 rounded-xl p-4"
+              >
+                <div className="flex items-center justify-between">
+                  <div className="flex items-center gap-3">
+                    {job.status === 'completed' && <CheckCircle className="text-success" size={18} />}
+                    {job.status === 'failed' && <AlertCircle className="text-error" size={18} />}
+                    {(job.status === 'pending' || job.status === 'running') && (
+                      <div className="animate-spin rounded-full h-[18px] w-[18px] border-t-2 border-b-2 border-secondary" />
+                    )}
+                    <span className="font-mono text-sm text-text-primary">{job.job_id.slice(0, 8)}</span>
+                    <span className="text-text-secondary text-sm">
+                      {job.total_companies} {job.total_companies === 1 ? 'company' : 'companies'}
+                    </span>
+                  </div>
+                  <span
+                    className={`text-xs font-semibold px-2.5 py-1 rounded-full ${
+                      job.status === 'completed'
+                        ? 'bg-success/15 text-success'
+                        : job.status === 'failed'
+                          ? 'bg-error/15 text-error'
+                          : 'bg-secondary/15 text-secondary'
+                    }`}
+                  >
+                    {job.status}
+                  </span>
+                </div>
+                {(job.status === 'running' || job.status === 'pending') && job.total_companies > 0 && (
+                  <div className="mt-3">
+                    <div className="flex items-center justify-between text-xs text-text-secondary mb-1">
+                      <span>Progress</span>
+                      <span>{job.completed_companies}/{job.total_companies}</span>
+                    </div>
+                    <div className="h-1.5 bg-bg-dark rounded-full overflow-hidden">
+                      <div
+                        className="h-full bg-gradient-to-r from-primary to-secondary rounded-full transition-all duration-300"
+                        style={{ width: `${(job.completed_companies / job.total_companies) * 100}%` }}
+                      />
+                    </div>
+                  </div>
+                )}
+                {job.status === 'failed' && job.error && (
+                  <p className="mt-2 text-sm text-error/80">{job.error}</p>
+                )}
+              </div>
+            ))}
+          </div>
+        )}
+      </div>
+    </div>
+  );
+}
+
+function SummaryCard({
+  label,
+  value,
+  color,
+}: {
+  label: string;
+  value: number | string;
+  color?: 'success' | 'error';
+}) {
+  const colorClass = color === 'success' ? 'text-success' : color === 'error' ? 'text-error' : '';
+
+  return (
+    <div className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-4 text-center">
+      <div
+        className={`text-2xl font-bold ${
+          colorClass || 'bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent'
+        }`}
+      >
+        {value}
+      </div>
+      <div className="text-sm text-text-secondary uppercase tracking-wide mt-1">{label}</div>
+    </div>
+  );
+}

frontend/src/pages/Compare.tsx

@@ -0,0 +1,161 @@
+import { useState } from 'react';
+import { useSearchParams } from 'react-router-dom';
+import { useQuery } from '@tanstack/react-query';
+import { analysisApi } from '../api/client';
+import { GitCompareArrows, AlertCircle, FileText, Clock } from 'lucide-react';
+import type { CompanyAnalysis } from '../types';
+
+function CompanyPanel({ data, isLoading, isError }: { data?: CompanyAnalysis; isLoading: boolean; isError: boolean }) {
+  if (isLoading) {
+    return (
+      <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6 animate-pulse">
+        <div className="h-6 w-32 bg-primary/20 rounded mb-4" />
+        <div className="space-y-3">
+          <div className="h-4 bg-primary/10 rounded w-full" />
+          <div className="h-4 bg-primary/10 rounded w-3/4" />
+          <div className="h-4 bg-primary/10 rounded w-5/6" />
+        </div>
+      </div>
+    );
+  }
+
+  if (isError) {
+    return (
+      <div className="bg-error/10 border border-error/20 rounded-2xl p-6">
+        <div className="flex items-center gap-2 text-error">
+          <AlertCircle size={18} />
+          <span>Failed to load analysis. Check the company name and try again.</span>
+        </div>
+      </div>
+    );
+  }
+
+  if (!data) return null;
+
+  return (
+    <div className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6 space-y-4">
+      <h3 className="text-lg font-bold text-text-primary border-b-2 border-primary/30 pb-2">
+        {data.company_name.toUpperCase()}
+      </h3>
+
+      <div className="grid grid-cols-2 gap-3">
+        <div className="bg-primary/10 rounded-lg p-3 text-center">
+          <FileText className="mx-auto mb-1 text-primary" size={18} />
+          <div className="text-xl font-bold text-text-primary">{data.patent_count}</div>
+          <div className="text-xs text-text-secondary uppercase">Patents</div>
+        </div>
+        <div className="bg-primary/10 rounded-lg p-3 text-center">
+          <Clock className="mx-auto mb-1 text-primary" size={18} />
+          <div className="text-sm font-medium text-text-primary">
+            {new Date(data.timestamp).toLocaleDateString()}
+          </div>
+          <div className="text-xs text-text-secondary uppercase">Analyzed</div>
+        </div>
+      </div>
+
+      {data.success && data.analysis ? (
+        <div className="text-text-primary whitespace-pre-wrap leading-relaxed text-sm">
+          {data.analysis}
+        </div>
+      ) : (
+        <div className="text-error text-sm">{data.error || 'Analysis not available'}</div>
+      )}
+    </div>
+  );
+}
+
+export function Compare() {
+  const [searchParams, setSearchParams] = useSearchParams();
+  const [companyA, setCompanyA] = useState(searchParams.get('a') || '');
+  const [companyB, setCompanyB] = useState(searchParams.get('b') || '');
+
+  const queryA = searchParams.get('a') || '';
+  const queryB = searchParams.get('b') || '';
+
+  const resultA = useQuery({
+    queryKey: ['analyze', queryA],
+    queryFn: () => analysisApi.analyzeCompany(queryA),
+    enabled: !!queryA,
+  });
+
+  const resultB = useQuery({
+    queryKey: ['analyze', queryB],
+    queryFn: () => analysisApi.analyzeCompany(queryB),
+    enabled: !!queryB,
+  });
+
+  const handleCompare = (e: React.FormEvent) => {
+    e.preventDefault();
+    const a = companyA.trim();
+    const b = companyB.trim();
+    if (a && b) {
+      setSearchParams({ a, b });
+    }
+  };
+
+  return (
+    <div className="space-y-6">
+      {/* Header */}
+      <div>
+        <h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
+          Portfolio Comparison
+        </h2>
+        <p className="text-text-secondary">
+          Compare patent portfolios of two companies side by side.
+        </p>
+      </div>
+
+      {/* Input Form */}
+      <form onSubmit={handleCompare} className="flex flex-col sm:flex-row gap-3 items-end">
+        <div className="flex-1">
+          <label className="block text-sm font-medium text-text-secondary mb-1">Company A</label>
+          <input
+            type="text"
+            value={companyA}
+            onChange={(e) => setCompanyA(e.target.value)}
+            placeholder="e.g. nvidia"
+            className="w-full bg-bg-card/80 border border-primary/30 rounded-xl px-4 py-2.5 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+          />
+        </div>
+        <div className="flex-1">
+          <label className="block text-sm font-medium text-text-secondary mb-1">Company B</label>
+          <input
+            type="text"
+            value={companyB}
+            onChange={(e) => setCompanyB(e.target.value)}
+            placeholder="e.g. intel"
+            className="w-full bg-bg-card/80 border border-primary/30 rounded-xl px-4 py-2.5 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+          />
+        </div>
+        <button
+          type="submit"
+          disabled={!companyA.trim() || !companyB.trim() || resultA.isLoading || resultB.isLoading}
+          className="bg-gradient-to-r from-primary to-primary-dark text-white font-semibold py-2.5 px-6 rounded-xl hover:shadow-lg hover:shadow-primary/30 transition-all disabled:opacity-50 disabled:cursor-not-allowed flex items-center gap-2"
+        >
+          <GitCompareArrows size={18} />
+          Compare
+        </button>
+      </form>
+
+      {/* Comparison Panels */}
+      {(queryA || queryB) && (
+        <div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
+          {queryA && (
+            <CompanyPanel
+              data={resultA.data}
+              isLoading={resultA.isLoading}
+              isError={resultA.isError}
+            />
+          )}
+          {queryB && (
+            <CompanyPanel
+              data={resultB.data}
+              isLoading={resultB.isLoading}
+              isError={resultB.isError}
+            />
+          )}
+        </div>
+      )}
+    </div>
+  );
+}

frontend/src/pages/Login.tsx

@@ -0,0 +1,121 @@
+import { useState } from 'react';
+import { Link, useNavigate, useLocation } from 'react-router-dom';
+import { useAuth } from '../context/AuthContext';
+import { LogIn, Mail, Lock, AlertCircle } from 'lucide-react';
+
+export function Login() {
+  const [email, setEmail] = useState('');
+  const [password, setPassword] = useState('');
+  const [error, setError] = useState('');
+  const [isLoading, setIsLoading] = useState(false);
+
+  const { login } = useAuth();
+  const navigate = useNavigate();
+  const location = useLocation();
+
+  const from = (location.state as { from?: { pathname: string } })?.from?.pathname || '/analysis';
+
+  const handleSubmit = async (e: React.FormEvent) => {
+    e.preventDefault();
+    setError('');
+    setIsLoading(true);
+
+    try {
+      await login(email, password);
+      navigate(from, { replace: true });
+    } catch (err) {
+      setError(err instanceof Error ? err.message : 'Invalid email or password');
+    } finally {
+      setIsLoading(false);
+    }
+  };
+
+  return (
+    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-slate-100 dark:to-indigo-950 flex items-center justify-center px-4">
+      <div className="w-full max-w-md">
+        {/* Brand */}
+        <div className="text-center mb-8">
+          <div className="flex items-center justify-center gap-3 mb-4">
+            <span className="text-4xl">⚡</span>
+            <h1 className="text-3xl font-bold bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent">
+              SPARC
+            </h1>
+          </div>
+          <p className="text-text-secondary">Semiconductor Patent Analytics Dashboard</p>
+        </div>
+
+        {/* Login Card */}
+        <div className="bg-bg-card/60 backdrop-blur-lg border border-primary/15 rounded-2xl p-8">
+          <h2 className="text-xl font-semibold text-text-primary mb-6">Sign in to your account</h2>
+
+          {error && (
+            <div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-lg px-4 py-3 mb-6">
+              <AlertCircle size={18} />
+              <span className="text-sm">{error}</span>
+            </div>
+          )}
+
+          <form onSubmit={handleSubmit} className="space-y-5">
+            <div>
+              <label htmlFor="email" className="block text-sm font-medium text-text-secondary mb-2">
+                Email
+              </label>
+              <div className="relative">
+                <Mail className="absolute left-3 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+                <input
+                  id="email"
+                  type="email"
+                  value={email}
+                  onChange={(e) => setEmail(e.target.value)}
+                  required
+                  className="w-full bg-bg-dark/80 border border-primary/30 rounded-xl pl-10 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+                  placeholder="you@example.com"
+                />
+              </div>
+            </div>
+
+            <div>
+              <label htmlFor="password" className="block text-sm font-medium text-text-secondary mb-2">
+                Password
+              </label>
+              <div className="relative">
+                <Lock className="absolute left-3 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+                <input
+                  id="password"
+                  type="password"
+                  value={password}
+                  onChange={(e) => setPassword(e.target.value)}
+                  required
+                  className="w-full bg-bg-dark/80 border border-primary/30 rounded-xl pl-10 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+                  placeholder="Enter your password"
+                />
+              </div>
+            </div>
+
+            <button
+              type="submit"
+              disabled={isLoading}
+              className="w-full bg-gradient-to-r from-primary to-primary-dark text-white font-semibold py-3 px-4 rounded-xl hover:shadow-lg hover:shadow-primary/30 transition-all disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
+            >
+              {isLoading ? (
+                <div className="animate-spin rounded-full h-5 w-5 border-t-2 border-b-2 border-white"></div>
+              ) : (
+                <>
+                  <LogIn size={18} />
+                  Sign In
+                </>
+              )}
+            </button>
+          </form>
+
+          <div className="mt-6 text-center">
+            <span className="text-text-secondary text-sm">Don't have an account? </span>
+            <Link to="/register" className="text-primary hover:text-primary-dark font-medium text-sm">
+              Sign up
+            </Link>
+          </div>
+        </div>
+      </div>
+    </div>
+  );
+}

frontend/src/pages/Register.tsx

@@ -0,0 +1,153 @@
+import { useState } from 'react';
+import { Link, useNavigate } from 'react-router-dom';
+import { useAuth } from '../context/AuthContext';
+import { UserPlus, Mail, Lock, AlertCircle } from 'lucide-react';
+
+export function Register() {
+  const [email, setEmail] = useState('');
+  const [password, setPassword] = useState('');
+  const [confirmPassword, setConfirmPassword] = useState('');
+  const [error, setError] = useState('');
+  const [isLoading, setIsLoading] = useState(false);
+
+  const { register } = useAuth();
+  const navigate = useNavigate();
+
+  const handleSubmit = async (e: React.FormEvent) => {
+    e.preventDefault();
+    setError('');
+
+    if (password !== confirmPassword) {
+      setError('Passwords do not match');
+      return;
+    }
+
+    if (password.length < 8) {
+      setError('Password must be at least 8 characters');
+      return;
+    }
+
+    setIsLoading(true);
+
+    try {
+      await register(email, password);
+      navigate('/analysis', { replace: true });
+    } catch (err) {
+      setError(err instanceof Error ? err.message : 'Registration failed');
+    } finally {
+      setIsLoading(false);
+    }
+  };
+
+  return (
+    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-slate-100 dark:to-indigo-950 flex items-center justify-center px-4">
+      <div className="w-full max-w-md">
+        {/* Brand */}
+        <div className="text-center mb-8">
+          <div className="flex items-center justify-center gap-3 mb-4">
+            <span className="text-4xl">⚡</span>
+            <h1 className="text-3xl font-bold bg-gradient-to-r from-primary to-secondary bg-clip-text text-transparent">
+              SPARC
+            </h1>
+          </div>
+          <p className="text-text-secondary">Semiconductor Patent Analytics Dashboard</p>
+        </div>
+
+        {/* Register Card */}
+        <div className="bg-bg-card/60 backdrop-blur-lg border border-primary/15 rounded-2xl p-8">
+          <h2 className="text-xl font-semibold text-text-primary mb-6">Create your account</h2>
+
+          {error && (
+            <div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-lg px-4 py-3 mb-6">
+              <AlertCircle size={18} />
+              <span className="text-sm">{error}</span>
+            </div>
+          )}
+
+          <form onSubmit={handleSubmit} className="space-y-5">
+            <div>
+              <label htmlFor="email" className="block text-sm font-medium text-text-secondary mb-2">
+                Email
+              </label>
+              <div className="relative">
+                <Mail className="absolute left-3 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+                <input
+                  id="email"
+                  type="email"
+                  value={email}
+                  onChange={(e) => setEmail(e.target.value)}
+                  required
+                  className="w-full bg-bg-dark/80 border border-primary/30 rounded-xl pl-10 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+                  placeholder="you@example.com"
+                />
+              </div>
+            </div>
+
+            <div>
+              <label htmlFor="password" className="block text-sm font-medium text-text-secondary mb-2">
+                Password
+              </label>
+              <div className="relative">
+                <Lock className="absolute left-3 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+                <input
+                  id="password"
+                  type="password"
+                  value={password}
+                  onChange={(e) => setPassword(e.target.value)}
+                  required
+                  minLength={8}
+                  className="w-full bg-bg-dark/80 border border-primary/30 rounded-xl pl-10 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+                  placeholder="At least 8 characters"
+                />
+              </div>
+            </div>
+
+            <div>
+              <label htmlFor="confirmPassword" className="block text-sm font-medium text-text-secondary mb-2">
+                Confirm Password
+              </label>
+              <div className="relative">
+                <Lock className="absolute left-3 top-1/2 -translate-y-1/2 text-text-secondary" size={18} />
+                <input
+                  id="confirmPassword"
+                  type="password"
+                  value={confirmPassword}
+                  onChange={(e) => setConfirmPassword(e.target.value)}
+                  required
+                  className="w-full bg-bg-dark/80 border border-primary/30 rounded-xl pl-10 pr-4 py-3 text-text-primary placeholder-text-secondary/50 focus:outline-none focus:border-primary focus:ring-2 focus:ring-primary/20 transition-all"
+                  placeholder="Confirm your password"
+                />
+              </div>
+            </div>
+
+            <button
+              type="submit"
+              disabled={isLoading}
+              className="w-full bg-gradient-to-r from-primary to-primary-dark text-white font-semibold py-3 px-4 rounded-xl hover:shadow-lg hover:shadow-primary/30 transition-all disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
+            >
+              {isLoading ? (
+                <div className="animate-spin rounded-full h-5 w-5 border-t-2 border-b-2 border-white"></div>
+              ) : (
+                <>
+                  <UserPlus size={18} />
+                  Create Account
+                </>
+              )}
+            </button>
+          </form>
+
+          <div className="mt-6 text-center">
+            <span className="text-text-secondary text-sm">Already have an account? </span>
+            <Link to="/login" className="text-primary hover:text-primary-dark font-medium text-sm">
+              Sign in
+            </Link>
+          </div>
+        </div>
+
+        <p className="mt-6 text-center text-xs text-text-secondary">
+          The first registered user will automatically become an admin.
+        </p>
+      </div>
+    </div>
+  );
+}

frontend/src/types/index.ts

@@ -0,0 +1,32 @@
+/**
+ * Application types derived from the auto-generated OpenAPI schema.
+ *
+ * Run `npm run generate:local` (or `npm run generate` with the API running)
+ * to regenerate `src/api/schema.d.ts` from the backend OpenAPI spec.
+ *
+ * These aliases keep the rest of the codebase stable while the source of
+ * truth lives in the generated file.
+ */
+
+import type { components } from '../api/schema';
+
+// Re-export schema types under the names the rest of the app expects.
+export type User = components['schemas']['UserResponse'];
+export type TokenResponse = components['schemas']['TokenResponse'];
+export type CompanyAnalysis = components['schemas']['CompanyAnalysisResponse'];
+export type BatchAnalysisResult = components['schemas']['BatchAnalysisResponse'];
+export type JobStatus = components['schemas']['JobStatus'];
+export type Analytics = Omit<components['schemas']['AnalyticsResponse'], 'by_company' | 'by_type'> & {
+  by_company: Array<{ company_name: string; count: number }>;
+  by_type: Array<{ analysis_type: string; count: number }>;
+};
+
+// Additional generated types that may be useful elsewhere.
+export type RegisterRequest = components['schemas']['RegisterRequest'];
+export type LoginRequest = components['schemas']['LoginRequest'];
+export type RefreshRequest = components['schemas']['RefreshRequest'];
+export type UpdateRoleRequest = components['schemas']['UpdateRoleRequest'];
+export type HealthResponse = components['schemas']['HealthResponse'];
+export type BatchAnalysisRequest = components['schemas']['BatchAnalysisRequest'];
+export type ValidationError = components['schemas']['ValidationError'];
+export type HTTPValidationError = components['schemas']['HTTPValidationError'];

frontend/src/vite-env.d.ts

@@ -0,0 +1,9 @@
+/// <reference types="vite/client" />
+
+interface ImportMetaEnv {
+  readonly VITE_API_URL: string;
+}
+
+interface ImportMeta {
+  readonly env: ImportMetaEnv;
+}

frontend/tailwind.config.js

@@ -0,0 +1,33 @@
+/** @type {import('tailwindcss').Config} */
+export default {
+  content: [
+    "./index.html",
+    "./src/**/*.{js,ts,jsx,tsx}",
+  ],
+  darkMode: 'class',
+  theme: {
+    extend: {
+      colors: {
+        primary: {
+          DEFAULT: '#6366f1',
+          dark: '#4f46e5',
+        },
+        secondary: '#0ea5e9',
+        success: '#10b981',
+        warning: '#f59e0b',
+        error: '#ef4444',
+        bg: {
+          dark: 'var(--color-bg-dark)',
+          card: 'var(--color-bg-card)',
+          'card-hover': 'var(--color-bg-card-hover)',
+        },
+        text: {
+          primary: 'var(--color-text-primary)',
+          secondary: 'var(--color-text-secondary)',
+        },
+        border: 'var(--color-border)',
+      },
+    },
+  },
+  plugins: [],
+}

frontend/tsconfig.json

@@ -0,0 +1,24 @@
+{
+  "compilerOptions": {
+    "target": "ES2020",
+    "useDefineForClassFields": true,
+    "lib": ["ES2020", "DOM", "DOM.Iterable"],
+    "module": "ESNext",
+    "skipLibCheck": true,
+    "moduleResolution": "bundler",
+    "allowImportingTsExtensions": true,
+    "isolatedModules": true,
+    "moduleDetection": "force",
+    "noEmit": true,
+    "jsx": "react-jsx",
+    "strict": true,
+    "noUnusedLocals": true,
+    "noUnusedParameters": true,
+    "noFallthroughCasesInSwitch": true,
+    "baseUrl": ".",
+    "paths": {
+      "@/*": ["src/*"]
+    }
+  },
+  "include": ["src"]
+}

frontend/vite.config.ts

@@ -0,0 +1,16 @@
+import { defineConfig } from 'vite'
+import react from '@vitejs/plugin-react'
+
+export default defineConfig({
+  plugins: [react()],
+  server: {
+    port: 3000,
+    proxy: {
+      '/api': {
+        target: 'http://localhost:8000',
+        changeOrigin: true,
+        rewrite: (path) => path.replace(/^\/api/, ''),
+      },
+    },
+  },
+})

main.py

@@ -1,10 +1,43 @@
-from SPARC.serp_api import SERP
+"""SPARC - Semiconductor Patent & Analytics Report Core
 
-patents = SERP.query("nvidia")
+Example usage of the company performance analyzer.
 
-for patent in patents.patents:
-  patent = SERP.save_patents(patent)
-  patent.summary = SERP.parse_patent_pdf(patent.pdf_path)
-  print(patent.summary)
+Before running:
+1. Create a .env file with:
+   API_KEY=your_serpapi_key
+   OPENROUTER_API_KEY=your_openrouter_key
 
-print(patents)
+2. Run: python main.py
+"""
+
+from SPARC.analyzer import CompanyAnalyzer
+
+
+def main():
+    """Analyze a company's performance based on their patent portfolio."""
+
+    # Initialize the analyzer (loads API keys from .env)
+    analyzer = CompanyAnalyzer()
+
+    # Analyze a company - this will:
+    # 1. Retrieve patents from SERP API
+    # 2. Download and parse patent PDFs
+    # 3. Minimize content (remove bloat)
+    # 4. Analyze with Claude to estimate performance
+    company_name = "nvidia"
+
+    print(f"\n{'=' * 70}")
+    print(f"SPARC Patent Analysis - {company_name.upper()}")
+    print(f"{'=' * 70}\n")
+
+    analysis = analyzer.analyze_company(company_name)
+
+    print(f"\n{'=' * 70}")
+    print("ANALYSIS RESULTS")
+    print(f"{'=' * 70}\n")
+    print(analysis)
+    print(f"\n{'=' * 70}\n")
+
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -4,4 +4,17 @@ pdfplumber
 requests
 pytest
 pytest-mock
-anthropic
+openai
+psycopg2-binary
+fastapi
+uvicorn[standard]
+pydantic[email]
+httpx
+numpy
+pandas
+bcrypt
+PyJWT
+slowapi
+apscheduler
+boto3
+reportlab

ruff.toml

@@ -0,0 +1,8 @@
+[lint]
+select = ["E", "F", "I"]
+ignore = [
+    "E501",  # line too long (handled by formatter)
+]
+
+[lint.per-file-ignores]
+"tests/*" = ["E402", "F841"]  # allow import not at top of file, unused vars (mocks) in tests

scripts/estimate_tokens.py

@@ -0,0 +1,227 @@
+#!/usr/bin/env python3
+"""Estimate token usage per company portfolio for SPARC analysis."""
+
+import tiktoken
+from typing import Dict, List, Optional
+from dataclasses import dataclass
+
+
+@dataclass
+class TokenEstimate:
+    """Token usage estimate for a company portfolio."""
+    company_name: str
+    patent_count: int
+    prompt_tokens: int
+    estimated_completion_tokens: int
+    total_tokens: int
+    cost_estimate_usd: float
+
+
+class TokenEstimator:
+    """Estimate token usage for SPARC patent analysis."""
+
+    # Claude 3.5 Sonnet pricing via OpenRouter (per 1M tokens)
+    INPUT_COST_PER_1M = 3.00   # $3.00 per 1M input tokens
+    OUTPUT_COST_PER_1M = 15.00  # $15.00 per 1M output tokens
+
+    # Estimated output tokens based on max_tokens settings
+    SINGLE_PATENT_MAX_OUTPUT = 1024
+    PORTFOLIO_MAX_OUTPUT = 2048
+
+    def __init__(self):
+        # Use cl100k_base encoding (closest to Claude's tokenizer)
+        self.encoder = tiktoken.get_encoding("cl100k_base")
+
+    def count_tokens(self, text: str) -> int:
+        """Count tokens in a text string."""
+        return len(self.encoder.encode(text))
+
+    def build_single_patent_prompt(self, patent_content: str, company_name: str) -> str:
+        """Build prompt for single patent analysis (matches llm.py)."""
+        return f"""You are a patent analyst evaluating {company_name}'s innovation strategy.
+
+Analyze the following patent content and provide insights on:
+1. Innovation quality and novelty
+2. Technical complexity and defensibility
+3. Market potential and commercial viability
+4. Strategic positioning relative to industry trends
+
+Patent Content:
+{patent_content}
+
+Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals about the company's technical direction and competitive advantage."""
+
+    def build_portfolio_prompt(self, patents_data: List[Dict[str, str]], company_name: str) -> str:
+        """Build prompt for portfolio analysis (matches llm.py)."""
+        portfolio_summary = []
+        for idx, patent in enumerate(patents_data, 1):
+            portfolio_summary.append(
+                f"Patent {idx} ({patent['patent_id']}):\n{patent['content']}"
+            )
+        combined_content = "\n\n---\n\n".join(portfolio_summary)
+
+        return f"""You are analyzing {company_name}'s patent portfolio to estimate their future performance and innovation trajectory.
+
+You have {len(patents_data)} recent patents to analyze. Evaluate the portfolio holistically:
+
+1. Innovation Trends: What technology areas are they focusing on?
+2. Strategic Direction: What does this reveal about their business strategy?
+3. Competitive Position: How defensible are these innovations?
+4. Market Outlook: What market opportunities do these patents target?
+5. Performance Forecast: Based on this innovation activity, what's your assessment of their likely performance?
+
+Patent Portfolio:
+{combined_content}
+
+Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the company's innovation strength and performance outlook."""
+
+    def estimate_portfolio(
+        self,
+        company_name: str,
+        patents_data: List[Dict[str, str]],
+        include_individual_patents: bool = False
+    ) -> TokenEstimate:
+        """Estimate tokens for a company portfolio analysis.
+
+        Args:
+            company_name: Name of the company
+            patents_data: List of dicts with 'patent_id' and 'content' keys
+            include_individual_patents: If True, also count individual patent analysis calls
+        """
+        # Portfolio analysis tokens
+        portfolio_prompt = self.build_portfolio_prompt(patents_data, company_name)
+        prompt_tokens = self.count_tokens(portfolio_prompt)
+        completion_tokens = self.PORTFOLIO_MAX_OUTPUT
+
+        # Optionally add individual patent analysis
+        if include_individual_patents:
+            for patent in patents_data:
+                single_prompt = self.build_single_patent_prompt(patent['content'], company_name)
+                prompt_tokens += self.count_tokens(single_prompt)
+                completion_tokens += self.SINGLE_PATENT_MAX_OUTPUT
+
+        total_tokens = prompt_tokens + completion_tokens
+
+        # Calculate cost
+        input_cost = (prompt_tokens / 1_000_000) * self.INPUT_COST_PER_1M
+        output_cost = (completion_tokens / 1_000_000) * self.OUTPUT_COST_PER_1M
+        total_cost = input_cost + output_cost
+
+        return TokenEstimate(
+            company_name=company_name,
+            patent_count=len(patents_data),
+            prompt_tokens=prompt_tokens,
+            estimated_completion_tokens=completion_tokens,
+            total_tokens=total_tokens,
+            cost_estimate_usd=total_cost
+        )
+
+    def estimate_from_sample(
+        self,
+        company_name: str,
+        patent_count: int = 10,
+        avg_patent_chars: int = 5000
+    ) -> TokenEstimate:
+        """Estimate tokens using sample/average patent sizes.
+
+        Args:
+            company_name: Name of the company
+            patent_count: Number of patents (default 10, typical from SERP)
+            avg_patent_chars: Average characters per minimized patent content
+        """
+        # Generate sample patent data
+        sample_content = "A" * avg_patent_chars  # Placeholder content
+        patents_data = [
+            {"patent_id": f"US{10000000 + i}", "content": sample_content}
+            for i in range(patent_count)
+        ]
+
+        return self.estimate_portfolio(company_name, patents_data)
+
+
+def main():
+    """Run token estimation examples."""
+    estimator = TokenEstimator()
+
+    print("=" * 70)
+    print("SPARC Token Usage Estimator")
+    print("=" * 70)
+
+    # Example 1: Estimate with sample data
+    print("\n📊 Sample Estimates (10 patents, ~5000 chars each):\n")
+
+    companies = ["Apple Inc.", "Microsoft Corporation", "Tesla Motors", "Google LLC"]
+
+    total_tokens = 0
+    total_cost = 0.0
+
+    for company in companies:
+        estimate = estimator.estimate_from_sample(company, patent_count=10, avg_patent_chars=5000)
+        print(f"  {company}:")
+        print(f"    Patents: {estimate.patent_count}")
+        print(f"    Prompt tokens: {estimate.prompt_tokens:,}")
+        print(f"    Est. completion tokens: {estimate.estimated_completion_tokens:,}")
+        print(f"    Total tokens: {estimate.total_tokens:,}")
+        print(f"    Est. cost: ${estimate.cost_estimate_usd:.4f}")
+        print()
+
+        total_tokens += estimate.total_tokens
+        total_cost += estimate.cost_estimate_usd
+
+    print("-" * 70)
+    print(f"  TOTAL for {len(companies)} companies:")
+    print(f"    Total tokens: {total_tokens:,}")
+    print(f"    Total est. cost: ${total_cost:.4f}")
+
+    # Example 2: Different portfolio sizes
+    print("\n" + "=" * 70)
+    print("📈 Token Scaling by Portfolio Size:")
+    print("=" * 70 + "\n")
+
+    for patent_count in [5, 10, 15, 20]:
+        estimate = estimator.estimate_from_sample("Sample Corp", patent_count=patent_count)
+        print(f"  {patent_count} patents: {estimate.prompt_tokens:,} prompt tokens, ${estimate.cost_estimate_usd:.4f}")
+
+    # Example 3: With actual patent content (simulated)
+    print("\n" + "=" * 70)
+    print("📝 Example with Real Patent Structure:")
+    print("=" * 70 + "\n")
+
+    sample_patents = [
+        {
+            "patent_id": "US11234567",
+            "content": """ABSTRACT: A method for machine learning optimization using gradient descent.
+
+CLAIMS:
+1. A computer-implemented method comprising:
+   receiving input data;
+   processing the input data through a neural network;
+   optimizing weights using backpropagation.
+
+SUMMARY: This invention relates to improvements in neural network training efficiency."""
+        },
+        {
+            "patent_id": "US11234568",
+            "content": """ABSTRACT: System for distributed computing in cloud environments.
+
+CLAIMS:
+1. A distributed system comprising:
+   a plurality of compute nodes;
+   a load balancer;
+   a message queue for task distribution.
+
+SUMMARY: The present disclosure improves cloud computing resource allocation."""
+        }
+    ]
+
+    estimate = estimator.estimate_portfolio("Tech Corp", sample_patents)
+    print(f"  Company: {estimate.company_name}")
+    print(f"  Patents analyzed: {estimate.patent_count}")
+    print(f"  Prompt tokens: {estimate.prompt_tokens:,}")
+    print(f"  Est. completion: {estimate.estimated_completion_tokens:,}")
+    print(f"  Total: {estimate.total_tokens:,}")
+    print(f"  Est. cost: ${estimate.cost_estimate_usd:.4f}")
+
+
+if __name__ == "__main__":
+    main()

scripts/init_database.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+"""Initialize the SPARC database schema.
+
+This script creates the necessary tables and indexes for storing
+LLM messages for testing and analytics.
+
+Usage:
+    python scripts/init_database.py
+"""
+
+import secrets
+import string
+import sys
+import os
+
+# Add parent directory to path
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from SPARC import config
+from SPARC.database import DatabaseClient
+
+DEFAULT_ADMIN_EMAIL = "admin@sparc.dev"
+
+
+def generate_password(length: int = 16) -> str:
+    """Generate a secure random password."""
+    alphabet = string.ascii_letters + string.digits
+    return "".join(secrets.choice(alphabet) for _ in range(length))
+
+
+def main():
+    """Initialize the database schema."""
+    print("Initializing SPARC database...")
+    print(f"Database URL: {config.database_url}")
+
+    try:
+        db_client = DatabaseClient(config.database_url)
+        db_client.initialize_schema()
+        print("Database schema initialized successfully!")
+        print("\nTables created:")
+        print("  - llm_messages: Stores all LLM prompts and responses")
+        print("  - users: Stores user accounts")
+        print("  - jobs: Stores async batch job state")
+        print("  - patents: Patent PDF cache")
+        print("  - serp_queries: SERP query result cache")
+        print("\nIndexes created:")
+        print("  - idx_messages_timestamp: For time-based queries")
+        print("  - idx_messages_company: For company-specific queries")
+        print("  - idx_users_email: For user lookups")
+
+        # Create default admin user if not exists
+        existing_admin = db_client.get_user_by_email(DEFAULT_ADMIN_EMAIL)
+        if existing_admin:
+            print(f"\nDefault admin user already exists: {DEFAULT_ADMIN_EMAIL}")
+        else:
+            password = generate_password()
+            admin_user = db_client.create_user(
+                email=DEFAULT_ADMIN_EMAIL,
+                password=password,
+                role="admin",
+            )
+            if admin_user:
+                print("\n" + "=" * 50)
+                print("DEFAULT ADMIN CREDENTIALS")
+                print("=" * 50)
+                print(f"Email:    {DEFAULT_ADMIN_EMAIL}")
+                print(f"Password: {password}")
+                print("=" * 50)
+                print("Please save these credentials securely!")
+                print("=" * 50)
+
+    except Exception as e:
+        print(f"Error initializing database: {e}")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

scripts/view_analytics.py

@@ -0,0 +1,67 @@
+#!/usr/bin/env python3
+"""View analytics from the message database.
+
+This script displays statistics about stored LLM messages including
+usage by company, analysis type, and time periods.
+
+Usage:
+    python scripts/view_analytics.py [--days DAYS]
+"""
+
+import sys
+import os
+import argparse
+import json
+
+# Add parent directory to path
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from SPARC import config
+from SPARC.database import DatabaseClient
+
+
+def main():
+    """Display analytics from the database."""
+    parser = argparse.ArgumentParser(description="View SPARC message analytics")
+    parser.add_argument(
+        "--days",
+        type=int,
+        default=30,
+        help="Number of days to analyze (default: 30)"
+    )
+    args = parser.parse_args()
+
+    print(f"SPARC Analytics - Last {args.days} days")
+    print("=" * 70)
+
+    try:
+        db_client = DatabaseClient(config.database_url)
+        analytics = db_client.get_analytics(days=args.days)
+
+        print(f"\nTotal Messages: {analytics['total_messages']}")
+
+        print("\nMessages by Company:")
+        if analytics['by_company']:
+            for item in analytics['by_company']:
+                company = item['company_name'] or '(unknown)'
+                print(f"  {company}: {item['count']}")
+        else:
+            print("  No data")
+
+        print("\nMessages by Analysis Type:")
+        if analytics['by_type']:
+            for item in analytics['by_type']:
+                analysis_type = item['analysis_type'] or '(unknown)'
+                print(f"  {analysis_type}: {item['count']}")
+        else:
+            print("  No data")
+
+        print("\n" + "=" * 70)
+
+    except Exception as e:
+        print(f"Error retrieving analytics: {e}")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

scripts/view_messages.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+"""View stored messages from the database.
+
+This script displays stored LLM messages with filtering options.
+
+Usage:
+    python scripts/view_messages.py [--company COMPANY] [--type TYPE] [--limit LIMIT]
+"""
+
+import sys
+import os
+import argparse
+from datetime import datetime
+
+# Add parent directory to path
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from SPARC import config
+from SPARC.database import DatabaseClient
+
+
+def main():
+    """Display messages from the database."""
+    parser = argparse.ArgumentParser(description="View stored SPARC messages")
+    parser.add_argument(
+        "--company",
+        type=str,
+        help="Filter by company name"
+    )
+    parser.add_argument(
+        "--type",
+        type=str,
+        choices=["single_patent", "portfolio"],
+        help="Filter by analysis type"
+    )
+    parser.add_argument(
+        "--limit",
+        type=int,
+        default=10,
+        help="Maximum number of messages to display (default: 10)"
+    )
+    args = parser.parse_args()
+
+    print("SPARC Stored Messages")
+    print("=" * 70)
+
+    try:
+        db_client = DatabaseClient(config.database_url)
+        messages = db_client.get_messages(
+            company_name=args.company,
+            analysis_type=args.type,
+            limit=args.limit
+        )
+
+        if not messages:
+            print("\nNo messages found.")
+            return
+
+        print(f"\nShowing {len(messages)} message(s):\n")
+
+        for i, msg in enumerate(messages, 1):
+            print(f"Message #{msg['id']} - {msg['timestamp']}")
+            print(f"Company: {msg['company_name'] or '(unknown)'}")
+            print(f"Type: {msg['analysis_type'] or '(unknown)'}")
+            print(f"Model: {msg['model'] or '(unknown)'}")
+            print(f"\nPrompt (first 200 chars):")
+            print(f"  {msg['prompt'][:200]}...")
+            print(f"\nResponse (first 200 chars):")
+            print(f"  {msg['response'][:200] if msg['response'] else '(no response)'}...")
+            print("\n" + "-" * 70 + "\n")
+
+    except Exception as e:
+        print(f"Error retrieving messages: {e}")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

test_database_mode.py

@@ -0,0 +1,141 @@
+#!/usr/bin/env python3
+"""Test script to verify database caching functionality.
+
+This script tests the LLMAnalyzer with database caching without requiring
+actual API keys or patent downloads.
+"""
+
+from SPARC.llm import LLMAnalyzer
+from SPARC.database import DatabaseClient
+from SPARC import config
+
+def test_database_storage():
+    """Test that messages are always stored in database."""
+    print("Testing Database Storage & Caching")
+    print("=" * 70)
+
+    # Initialize analyzer (database is always used)
+    print("\n1. Initializing LLMAnalyzer...")
+    analyzer = LLMAnalyzer(use_cache=True)
+
+    print(f"   - use_cache: {analyzer.use_cache}")
+    print(f"   - db_client: {analyzer.db_client is not None}")
+    print(f"   - client (API): {analyzer.client is not None}")
+
+    # Test single patent analysis (without API key, stores placeholder)
+    print("\n2. Testing single patent analysis (no API key)...")
+    result = analyzer.analyze_patent_content(
+        patent_content="Test patent content about semiconductor innovation",
+        company_name="TestCorp"
+    )
+    print(f"   Result: {result[:80]}...")
+
+    # Test portfolio analysis
+    print("\n3. Testing portfolio analysis (no API key)...")
+    test_patents = [
+        {"patent_id": "US001", "content": "First test patent"},
+        {"patent_id": "US002", "content": "Second test patent"},
+    ]
+    result = analyzer.analyze_patent_portfolio(
+        patents_data=test_patents,
+        company_name="TestCorp"
+    )
+    print(f"   Result: {result[:80]}...")
+
+    # Verify messages were stored
+    print("\n4. Verifying messages were stored...")
+    db_client = DatabaseClient(config.database_url)
+    messages = db_client.get_messages(company_name="TestCorp", limit=10)
+    print(f"   Found {len(messages)} stored messages")
+
+    for msg in messages:
+        cached_status = "CACHED" if msg.get('is_cached') else "NEW"
+        print(f"   - ID: {msg['id']}, Type: {msg['analysis_type']}, Status: {cached_status}")
+
+    # Get analytics
+    print("\n5. Getting analytics...")
+    analytics = db_client.get_analytics(days=1)
+    print(f"   Total messages: {analytics['total_messages']}")
+    print(f"   By company: {analytics['by_company']}")
+    print(f"   By type: {analytics['by_type']}")
+
+    print("\n" + "=" * 70)
+    print("Database storage test completed successfully!")
+
+def test_caching():
+    """Test that caching works correctly."""
+    print("\nTesting Cache Functionality")
+    print("=" * 70)
+
+    db_client = DatabaseClient(config.database_url)
+    db_client.initialize_schema()
+
+    # Store a fake cached response
+    print("\n1. Storing a test response in database...")
+    test_prompt = "Test prompt for caching"
+    test_response = "This is a cached response from previous API call"
+
+    db_client.store_message(
+        prompt=test_prompt,
+        response=test_response,
+        company_name="CacheTest",
+        analysis_type="test",
+        model="test-model"
+    )
+
+    # Try to retrieve from cache
+    print("\n2. Testing cache retrieval...")
+    cached = db_client.get_cached_response(
+        prompt=test_prompt,
+        company_name="CacheTest",
+        analysis_type="test"
+    )
+
+    if cached:
+        print(f"   Cache hit! Response: {cached['response']}")
+    else:
+        print("   Cache miss (unexpected)")
+
+    # Test cache miss
+    print("\n3. Testing cache miss...")
+    cached = db_client.get_cached_response(
+        prompt="Different prompt",
+        company_name="CacheTest",
+        analysis_type="test"
+    )
+
+    if cached:
+        print("   Unexpected cache hit")
+    else:
+        print("   Cache miss as expected")
+
+    print("\n" + "=" * 70)
+    print("Cache test completed successfully!")
+
+def test_test_mode():
+    """Test that test mode works correctly."""
+    print("\nTesting Test Mode")
+    print("=" * 70)
+
+    print("\n1. Initializing LLMAnalyzer in test mode...")
+    analyzer = LLMAnalyzer(test_mode=True)
+
+    print(f"   - test_mode: {analyzer.test_mode}")
+    print(f"   - db_client: {analyzer.db_client is not None}")
+
+    print("\n2. Testing single patent analysis (test mode)...")
+    result = analyzer.analyze_patent_content(
+        patent_content="Test patent content",
+        company_name="TestCorp2"
+    )
+    print(f"   Result: {result}")
+
+    print("\n" + "=" * 70)
+    print("Test mode test completed successfully!")
+
+if __name__ == "__main__":
+    test_database_storage()
+    print("\n")
+    test_caching()
+    print("\n")
+    test_test_mode()

tests/test_analyzer.py

@@ -0,0 +1,542 @@
+"""Tests for the high-level company analyzer orchestration."""
+
+from unittest.mock import MagicMock, Mock
+
+import pytest
+
+from SPARC.analyzer import CompanyAnalyzer
+from SPARC.types import BatchAnalysisResult, Patent, Patents
+
+
+@pytest.fixture(autouse=True)
+def mock_db(mocker):
+    """Mock DatabaseClient for all tests so no real DB connection is needed."""
+    mock_db_cls = mocker.patch("SPARC.analyzer.DatabaseClient")
+    mock_db_instance = MagicMock()
+    mock_db_instance.get_cached_patent.return_value = None
+    mock_db_instance.get_cached_serp_query.return_value = None
+    mock_db_cls.return_value = mock_db_instance
+    return mock_db_instance
+
+
+class TestCompanyAnalyzer:
+    """Test the CompanyAnalyzer orchestration logic."""
+
+    def test_analyzer_initialization(self, mocker):
+        """Test analyzer initialization with API key."""
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        _analyzer = CompanyAnalyzer(openrouter_api_key="test-key")  # noqa: F841
+
+        mock_llm.assert_called_once_with(api_key="test-key")
+
+    def test_analyze_company_full_pipeline(self, mocker, mock_db):
+        """Test complete company analysis pipeline."""
+        # Mock all the dependencies
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        # Setup mock return values
+        test_patent = Patent(
+            patent_id="US123", pdf_link="http://example.com/test.pdf"
+        )
+        mock_query.return_value = Patents(patents=[test_patent])
+
+        test_patent.pdf_path = "patents/US123.pdf"
+        mock_save.return_value = test_patent
+
+        mock_parse.return_value = {
+            "abstract": "Test abstract",
+            "claims": "Test claims",
+        }
+
+        mock_minimize.return_value = "Minimized content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = (
+            "Strong innovation portfolio"
+        )
+        mock_llm.return_value = mock_llm_instance
+
+        # Run the analysis
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_company("TestCorp")
+
+        # Verify the pipeline executed correctly
+        assert result == "Strong innovation portfolio"
+        mock_query.assert_called_once_with("TestCorp")
+        mock_save.assert_called_once()
+        mock_parse.assert_called_once_with("patents/US123.pdf")
+        mock_minimize.assert_called_once()
+        mock_llm_instance.analyze_patent_portfolio.assert_called_once()
+
+        # Verify the data passed to LLM
+        llm_call_args = mock_llm_instance.analyze_patent_portfolio.call_args
+        patents_data = llm_call_args[1]["patents_data"]
+        assert len(patents_data) == 1
+        assert patents_data[0]["patent_id"] == "US123"
+        assert patents_data[0]["content"] == "Minimized content"
+
+    def test_analyze_company_no_patents_found(self, mocker):
+        """Test handling when no patents are found for a company."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_query.return_value = Patents(patents=[])
+        mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_company("UnknownCorp")
+
+        assert result == "No patents found for UnknownCorp"
+
+    def test_analyze_company_handles_processing_errors(self, mocker):
+        """Test that analysis continues even if some patents fail to process."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        # Create two test patents
+        patent1 = Patent(patent_id="US123", pdf_link="http://example.com/1.pdf")
+        patent2 = Patent(patent_id="US456", pdf_link="http://example.com/2.pdf")
+        mock_query.return_value = Patents(patents=[patent1, patent2])
+
+        # First patent processes successfully
+        patent1.pdf_path = "patents/US123.pdf"
+
+        # Second patent raises an error
+        def save_side_effect(p):
+            if p.patent_id == "US123":
+                p.pdf_path = "patents/US123.pdf"
+                return p
+            else:
+                raise Exception("Download failed")
+
+        mock_save.side_effect = save_side_effect
+
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis result"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_company("TestCorp")
+
+        # Should still succeed with the one patent that worked
+        assert result == "Analysis result"
+
+        # Verify only one patent was analyzed
+        llm_call_args = mock_llm_instance.analyze_patent_portfolio.call_args
+        patents_data = llm_call_args[1]["patents_data"]
+        assert len(patents_data) == 1
+        assert patents_data[0]["patent_id"] == "US123"
+
+    def test_analyze_company_all_patents_fail(self, mocker):
+        """Test handling when all patents fail to process."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/1.pdf")
+        mock_query.return_value = Patents(patents=[patent])
+
+        # Make processing fail
+        mock_save.side_effect = Exception("Processing error")
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_company("TestCorp")
+
+        assert result == "Failed to process any patents for TestCorp"
+
+    def test_analyze_single_patent(self, mocker):
+        """Test single patent analysis."""
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        mock_parse.return_value = {"abstract": "Test abstract"}
+        mock_minimize.return_value = "Minimized content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_content.return_value = (
+            "Innovative patent analysis"
+        )
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_single_patent("US123", "TestCorp")
+
+        assert result == "Innovative patent analysis"
+        mock_parse.assert_called_once_with("patents/US123.pdf")
+        mock_llm_instance.analyze_patent_content.assert_called_once_with(
+            patent_content="Minimized content", company_name="TestCorp"
+        )
+
+    def test_analyze_single_patent_error_handling(self, mocker):
+        """Test single patent analysis with processing error."""
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        mock_parse.side_effect = FileNotFoundError("PDF not found")
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_single_patent("US999", "TestCorp")
+
+        assert "Failed to analyze patent US999" in result
+        assert "PDF not found" in result
+
+
+class TestSingleQueryBugFix:
+    """Test that SERP.query is only called once per company analysis."""
+
+    def test_analyze_company_safe_calls_query_once(self, mocker, mock_db):
+        """_analyze_company_safe should call SERP.query exactly once."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        mock_query.return_value = Patents(patents=[patent])
+
+        def save_side_effect(p):
+            p.pdf_path = "patents/US123.pdf"
+            return p
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        analyzer._analyze_company_safe("TestCorp")
+
+        # The key assertion: SERP.query called exactly once, not twice
+        mock_query.assert_called_once_with("TestCorp")
+
+    def test_analyze_company_with_prefetched_patents_skips_query(self, mocker):
+        """analyze_company should not call SERP.query when patents are provided."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        prefetched = Patents(patents=[patent])
+
+        def save_side_effect(p):
+            p.pdf_path = "patents/US123.pdf"
+            return p
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        analyzer.analyze_company("TestCorp", patents=prefetched)
+
+        # SERP.query should never be called
+        mock_query.assert_not_called()
+
+
+class TestPatentCaching:
+    """Test patent-level DB caching in the pipeline."""
+
+    def test_process_single_patent_uses_db_cache(self, mocker, mock_db):
+        """_process_single_patent returns cached content when available."""
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+
+        mock_db.get_cached_patent.return_value = {
+            "patent_id": "US123",
+            "minimized_content": "Cached minimized content",
+        }
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        result = CompanyAnalyzer._process_single_patent(patent, "TestCorp", mock_db)
+
+        assert result == {"patent_id": "US123", "content": "Cached minimized content"}
+        # Should NOT download since cache hit
+        mock_save.assert_not_called()
+
+    def test_process_single_patent_stores_to_db_cache(self, mocker, mock_db):
+        """_process_single_patent stores result in DB after processing."""
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+
+        # No cache hit
+        mock_db.get_cached_patent.return_value = None
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+
+        def save_side_effect(p):
+            p.pdf_path = "patents/US123.pdf"
+            return p
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test abstract"}
+        mock_minimize.return_value = "Minimized content"
+
+        result = CompanyAnalyzer._process_single_patent(patent, "TestCorp", mock_db)
+
+        assert result == {"patent_id": "US123", "content": "Minimized content"}
+        mock_db.store_patent.assert_called_once_with(
+            patent_id="US123",
+            company_name="TestCorp",
+            pdf_link="http://example.com/test.pdf",
+            raw_sections={"abstract": "Test abstract"},
+            minimized_content="Minimized content",
+        )
+
+    def test_serp_query_cache_hit_skips_api(self, mocker, mock_db):
+        """When SERP query is cached, API call is skipped."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        # Simulate SERP cache hit
+        mock_db.get_cached_serp_query.return_value = ["US123"]
+        # Simulate patent cache hit too
+        mock_db.get_cached_patent.return_value = {
+            "patent_id": "US123",
+            "minimized_content": "Cached content",
+        }
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_company("TestCorp")
+
+        assert result == "Analysis"
+        # SERP.query should NOT be called
+        mock_query.assert_not_called()
+        # No downloads should happen
+        mock_save.assert_not_called()
+
+    def test_serp_query_cache_miss_stores_result(self, mocker, mock_db):
+        """When SERP query cache misses, result is stored after API call."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        mock_db.get_cached_serp_query.return_value = None
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        mock_query.return_value = Patents(patents=[patent])
+
+        def save_side_effect(p):
+            p.pdf_path = "patents/US123.pdf"
+            return p
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        analyzer.analyze_company("TestCorp")
+
+        mock_db.store_serp_query.assert_called_once()
+        call_kwargs = mock_db.store_serp_query.call_args[1]
+        assert call_kwargs["company_name"] == "TestCorp"
+        assert call_kwargs["patent_ids"] == ["US123"]
+
+
+class TestBatchProcessing:
+    """Test multi-company batch processing functionality."""
+
+    def test_analyze_companies_success(self, mocker):
+        """Test batch analysis of multiple companies."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        # Setup mock returns
+        def query_side_effect(company):
+            patent = Patent(
+                patent_id=f"US-{company}",
+                pdf_link=f"http://example.com/{company}.pdf",
+            )
+            return Patents(patents=[patent])
+
+        mock_query.side_effect = query_side_effect
+
+        def save_side_effect(patent):
+            patent.pdf_path = f"patents/{patent.patent_id}.pdf"
+            return patent
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis result"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_companies(["CompanyA", "CompanyB"], max_workers=2)
+
+        assert isinstance(result, BatchAnalysisResult)
+        assert result.total_companies == 2
+        assert result.successful == 2
+        assert result.failed == 0
+        assert len(result.results) == 2
+
+    def test_analyze_companies_with_failures(self, mocker):
+        """Test batch analysis handles partial failures."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        def query_side_effect(company):
+            if company == "FailCorp":
+                return Patents(patents=[])
+            patent = Patent(
+                patent_id=f"US-{company}",
+                pdf_link=f"http://example.com/{company}.pdf",
+            )
+            return Patents(patents=[patent])
+
+        mock_query.side_effect = query_side_effect
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_companies(["GoodCorp", "FailCorp"], max_workers=1)
+
+        assert result.total_companies == 2
+        assert result.failed >= 1  # At least FailCorp should fail
+
+        # Find the failed result
+        fail_result = next(r for r in result.results if r.company_name == "FailCorp")
+        assert fail_result.success is False
+
+    def test_analyze_companies_sequential(self, mocker):
+        """Test sequential batch analysis."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        def query_side_effect(company):
+            patent = Patent(
+                patent_id=f"US-{company}",
+                pdf_link=f"http://example.com/{company}.pdf",
+            )
+            return Patents(patents=[patent])
+
+        mock_query.side_effect = query_side_effect
+
+        def save_side_effect(patent):
+            patent.pdf_path = f"patents/{patent.patent_id}.pdf"
+            return patent
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_companies_sequential(["Corp1", "Corp2", "Corp3"])
+
+        assert result.total_companies == 3
+        assert len(result.results) == 3
+
+    def test_analyze_companies_progress_callback(self, mocker):
+        """Test that progress callback is invoked correctly."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        def query_side_effect(company):
+            patent = Patent(
+                patent_id=f"US-{company}",
+                pdf_link=f"http://example.com/{company}.pdf",
+            )
+            return Patents(patents=[patent])
+
+        mock_query.side_effect = query_side_effect
+
+        def save_side_effect(patent):
+            patent.pdf_path = f"patents/{patent.patent_id}.pdf"
+            return patent
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Analysis"
+        mock_llm.return_value = mock_llm_instance
+
+        callback = Mock()
+        analyzer = CompanyAnalyzer()
+        analyzer.analyze_companies(["A", "B"], max_workers=1, progress_callback=callback)
+
+        assert callback.call_count == 2
+
+    def test_company_analysis_result_structure(self, mocker, mock_db):
+        """Test CompanyAnalysisResult has correct structure."""
+        mock_query = mocker.patch("SPARC.analyzer.SERP.query")
+        mock_save = mocker.patch("SPARC.analyzer.SERP.save_patents")
+        mock_parse = mocker.patch("SPARC.analyzer.SERP.parse_patent_pdf")
+        mock_minimize = mocker.patch("SPARC.analyzer.SERP.minimize_patent_for_llm")
+        mock_llm = mocker.patch("SPARC.analyzer.LLMAnalyzer")
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        mock_query.return_value = Patents(patents=[patent])
+
+        # Simulate DB caching: after store, subsequent get returns the IDs
+        mock_db.get_cached_serp_query.side_effect = [None, ["US123"]]
+
+        def save_side_effect(p):
+            p.pdf_path = "patents/US123.pdf"
+            return p
+
+        mock_save.side_effect = save_side_effect
+        mock_parse.return_value = {"abstract": "Test"}
+        mock_minimize.return_value = "Content"
+
+        mock_llm_instance = Mock()
+        mock_llm_instance.analyze_patent_portfolio.return_value = "Strong innovation"
+        mock_llm.return_value = mock_llm_instance
+
+        analyzer = CompanyAnalyzer()
+        result = analyzer.analyze_companies(["TestCorp"], max_workers=1)
+
+        assert len(result.results) == 1
+        company_result = result.results[0]
+        assert company_result.company_name == "TestCorp"
+        assert company_result.analysis == "Strong innovation"
+        assert company_result.patent_count == 1
+        assert company_result.success is True
+        assert company_result.error is None
+        assert company_result.timestamp is not None

tests/test_api.py

@@ -0,0 +1,228 @@
+"""Tests for FastAPI web service endpoints."""
+
+from datetime import datetime
+from unittest.mock import Mock
+
+import pytest
+from fastapi.testclient import TestClient
+
+from SPARC.api import app
+from SPARC.types import BatchAnalysisResult, CompanyAnalysisResult
+
+
+@pytest.fixture
+def client():
+    """Create test client."""
+    return TestClient(app)
+
+
+@pytest.fixture
+def mock_analyzer(mocker):
+    """Mock the global analyzer."""
+    mock = Mock()
+    mocker.patch("SPARC.api._analyzer", mock)
+    return mock
+
+
+class TestHealthEndpoint:
+    """Test health check endpoint."""
+
+    def test_health_returns_ok(self, client):
+        """Test health endpoint returns healthy status."""
+        response = client.get("/health")
+        assert response.status_code == 200
+        data = response.json()
+        assert data["status"] == "healthy"
+        assert data["version"] == "1.0.0"
+        assert "timestamp" in data
+
+
+class TestAnalyzeCompanyEndpoint:
+    """Test single company analysis endpoint."""
+
+    def test_analyze_company_success(self, client, mock_analyzer):
+        """Test successful company analysis."""
+        mock_result = CompanyAnalysisResult(
+            company_name="nvidia",
+            analysis="Strong AI patent portfolio",
+            patent_count=5,
+            success=True,
+            timestamp=datetime.now(),
+        )
+        mock_analyzer._analyze_company_safe.return_value = mock_result
+
+        response = client.get("/analyze/nvidia")
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["company_name"] == "nvidia"
+        assert data["analysis"] == "Strong AI patent portfolio"
+        assert data["patent_count"] == 5
+        assert data["success"] is True
+
+    def test_analyze_company_failure(self, client, mock_analyzer):
+        """Test company analysis with error."""
+        mock_result = CompanyAnalysisResult(
+            company_name="unknown",
+            analysis="",
+            patent_count=0,
+            success=False,
+            error="No patents found",
+            timestamp=datetime.now(),
+        )
+        mock_analyzer._analyze_company_safe.return_value = mock_result
+
+        response = client.get("/analyze/unknown")
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["success"] is False
+        assert data["error"] == "No patents found"
+
+
+class TestBatchAnalysisEndpoint:
+    """Test batch analysis endpoint."""
+
+    def test_batch_analysis_success(self, client, mock_analyzer):
+        """Test successful batch analysis."""
+        results = [
+            CompanyAnalysisResult(
+                company_name="nvidia",
+                analysis="Strong portfolio",
+                patent_count=5,
+                success=True,
+                timestamp=datetime.now(),
+            ),
+            CompanyAnalysisResult(
+                company_name="amd",
+                analysis="Growing portfolio",
+                patent_count=3,
+                success=True,
+                timestamp=datetime.now(),
+            ),
+        ]
+        mock_batch = BatchAnalysisResult(
+            results=results,
+            total_companies=2,
+            successful=2,
+            failed=0,
+            timestamp=datetime.now(),
+        )
+        mock_analyzer.analyze_companies.return_value = mock_batch
+
+        response = client.post(
+            "/analyze/batch",
+            json={"companies": ["nvidia", "amd"], "max_workers": 2},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["total_companies"] == 2
+        assert data["successful"] == 2
+        assert data["failed"] == 0
+        assert len(data["results"]) == 2
+
+    def test_batch_analysis_validation(self, client):
+        """Test batch analysis request validation."""
+        # Empty companies list
+        response = client.post("/analyze/batch", json={"companies": []})
+        assert response.status_code == 422
+
+        # Too many companies
+        response = client.post(
+            "/analyze/batch",
+            json={"companies": [f"company{i}" for i in range(25)]},
+        )
+        assert response.status_code == 422
+
+        # Invalid max_workers
+        response = client.post(
+            "/analyze/batch",
+            json={"companies": ["nvidia"], "max_workers": 10},
+        )
+        assert response.status_code == 422
+
+
+class TestAsyncBatchEndpoint:
+    """Test async batch analysis endpoint."""
+
+    def test_async_batch_creates_job(self, client, mock_analyzer):
+        """Test async endpoint creates a job."""
+        response = client.post(
+            "/analyze/batch/async",
+            json={"companies": ["nvidia", "amd"]},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert "job_id" in data
+        assert data["status"] == "pending"
+        assert data["total_companies"] == 2
+        assert data["progress"] == 0
+
+
+class TestJobEndpoints:
+    """Test job management endpoints."""
+
+    def test_get_job_not_found(self, client):
+        """Test getting nonexistent job."""
+        response = client.get("/jobs/nonexistent")
+        assert response.status_code == 404
+
+    def test_list_jobs(self, client, mocker):
+        """Test listing jobs."""
+        # Clear existing jobs
+        mocker.patch.dict("SPARC.api._jobs", {}, clear=True)
+
+        response = client.get("/jobs")
+        assert response.status_code == 200
+        assert isinstance(response.json(), list)
+
+    def test_list_jobs_with_filter(self, client, mocker):
+        """Test listing jobs with status filter."""
+        response = client.get("/jobs?status=completed")
+        assert response.status_code == 200
+
+
+class TestModelValidation:
+    """Test that unsupported model identifiers are rejected."""
+
+    def test_analyze_rejects_unsupported_model(self, client, mock_analyzer):
+        """GET /analyze/{company} with unsupported model returns 400."""
+        response = client.get("/analyze/nvidia?model=fake/nonexistent-model")
+        assert response.status_code == 400
+        assert "Unsupported model" in response.json()["detail"]
+
+    def test_analyze_accepts_supported_model(self, client, mock_analyzer):
+        """GET /analyze/{company} with a supported model succeeds."""
+        mock_result = CompanyAnalysisResult(
+            company_name="nvidia",
+            analysis="test",
+            patent_count=1,
+            success=True,
+            timestamp=datetime.now(),
+            model="anthropic/claude-3.5-sonnet",
+        )
+        mock_analyzer._analyze_company_safe.return_value = mock_result
+
+        response = client.get("/analyze/nvidia?model=anthropic/claude-3.5-sonnet")
+        assert response.status_code == 200
+
+    def test_batch_rejects_unsupported_model(self, client, mock_analyzer):
+        """POST /analyze/batch with unsupported model returns 400."""
+        response = client.post(
+            "/analyze/batch",
+            json={"companies": ["nvidia"], "model": "fake/nonexistent-model"},
+        )
+        assert response.status_code == 400
+        assert "Unsupported model" in response.json()["detail"]
+
+    def test_list_models_returns_supported(self, client):
+        """GET /models returns the allow-list."""
+        response = client.get("/models")
+        assert response.status_code == 200
+        data = response.json()
+        assert "models" in data
+        assert "default" in data
+        assert len(data["models"]) > 0
+        assert all("id" in m and "name" in m and "provider" in m for m in data["models"])

tests/test_auth.py

@@ -0,0 +1,302 @@
+"""Tests for JWT authentication flow: register, login, protected routes, refresh, admin access."""
+
+from datetime import datetime, timezone
+from unittest.mock import MagicMock, patch
+
+import pytest
+from fastapi.testclient import TestClient
+
+from SPARC.api import app
+from SPARC.auth import create_access_token, create_refresh_token
+
+
+@pytest.fixture
+def client():
+    """Create test client."""
+    return TestClient(app)
+
+
+@pytest.fixture(autouse=True)
+def mock_db(monkeypatch):
+    """Mock the database client used by auth endpoints.
+
+    Returns a MagicMock with all DB methods pre-configured.
+    """
+    db = MagicMock()
+
+    # Default: no users exist
+    db.get_user_count.return_value = 0
+    db.get_user_by_id.return_value = None
+    db.get_user_by_email.return_value = None
+    db.authenticate_user.return_value = None
+    db.create_user.return_value = None
+    db.get_all_users.return_value = []
+    db.update_user_role.return_value = None
+    db.delete_user.return_value = False
+
+    with patch("SPARC.api.get_db_client", return_value=db), \
+         patch("SPARC.auth.get_db_client", return_value=db):
+        yield db
+
+
+def _make_admin_user():
+    return {
+        "id": 1,
+        "email": "admin@test.com",
+        "role": "admin",
+        "created_at": datetime(2025, 1, 1, tzinfo=timezone.utc),
+    }
+
+
+def _make_regular_user():
+    return {
+        "id": 2,
+        "email": "user@test.com",
+        "role": "user",
+        "created_at": datetime(2025, 1, 1, tzinfo=timezone.utc),
+    }
+
+
+def _auth_header(user_dict):
+    """Create an Authorization header with a valid access token for the given user."""
+    token = create_access_token(user_dict["id"], user_dict["email"], user_dict["role"])
+    return {"Authorization": f"Bearer {token}"}
+
+
+class TestRegister:
+    """POST /auth/register"""
+
+    def test_register_first_user_becomes_admin(self, client, mock_db):
+        """First registered user should get admin role."""
+        mock_db.get_user_count.return_value = 0
+        mock_db.create_user.return_value = {
+            "id": 1,
+            "email": "admin@test.com",
+            "role": "admin",
+            "created_at": datetime(2025, 1, 1, tzinfo=timezone.utc),
+        }
+
+        response = client.post(
+            "/auth/register",
+            json={"email": "admin@test.com", "password": "securepass123"},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["email"] == "admin@test.com"
+        assert data["role"] == "admin"
+        mock_db.create_user.assert_called_once_with(
+            email="admin@test.com", password="securepass123", role="admin"
+        )
+
+    def test_register_subsequent_user_gets_user_role(self, client, mock_db):
+        """Non-first user should get regular user role."""
+        mock_db.get_user_count.return_value = 1
+        mock_db.create_user.return_value = _make_regular_user()
+
+        response = client.post(
+            "/auth/register",
+            json={"email": "user@test.com", "password": "securepass123"},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["role"] == "user"
+
+    def test_register_duplicate_email_returns_400(self, client, mock_db):
+        """Registering with an existing email should return 400."""
+        mock_db.get_user_count.return_value = 1
+        mock_db.create_user.return_value = None  # indicates duplicate
+
+        response = client.post(
+            "/auth/register",
+            json={"email": "existing@test.com", "password": "securepass123"},
+        )
+
+        assert response.status_code == 400
+        assert "already registered" in response.json()["detail"].lower()
+
+
+class TestLogin:
+    """POST /auth/login"""
+
+    def test_login_valid_credentials_returns_tokens(self, client, mock_db):
+        """Valid credentials should return access and refresh tokens."""
+        user = _make_regular_user()
+        mock_db.authenticate_user.return_value = user
+
+        response = client.post(
+            "/auth/login",
+            json={"email": "user@test.com", "password": "correctpassword"},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert "access_token" in data
+        assert "refresh_token" in data
+        assert data["token_type"] == "bearer"
+
+    def test_login_invalid_credentials_returns_401(self, client, mock_db):
+        """Invalid credentials should return 401."""
+        mock_db.authenticate_user.return_value = None
+
+        response = client.post(
+            "/auth/login",
+            json={"email": "user@test.com", "password": "wrongpassword"},
+        )
+
+        assert response.status_code == 401
+        assert "invalid" in response.json()["detail"].lower()
+
+
+class TestGetMe:
+    """GET /auth/me"""
+
+    def test_valid_access_token_returns_user(self, client, mock_db):
+        """A valid access token should return the user's data."""
+        user = _make_regular_user()
+        mock_db.get_user_by_id.return_value = user
+
+        response = client.get("/auth/me", headers=_auth_header(user))
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["email"] == "user@test.com"
+        assert data["id"] == 2
+
+    def test_missing_token_returns_401(self, client):
+        """No token should return 401 (403 from HTTPBearer)."""
+        response = client.get("/auth/me")
+        assert response.status_code in (401, 403)
+
+    def test_expired_token_returns_401(self, client, mock_db):
+        """An expired token should return 401."""
+        # Create a token that has already expired
+        from datetime import timedelta
+
+        import jwt as pyjwt
+        from SPARC.auth import JWT_ALGORITHM, JWT_SECRET
+
+        payload = {
+            "sub": "1",
+            "email": "user@test.com",
+            "role": "user",
+            "exp": datetime.now(timezone.utc) - timedelta(hours=1),
+            "type": "access",
+        }
+        expired_token = pyjwt.encode(payload, JWT_SECRET, algorithm=JWT_ALGORITHM)
+
+        response = client.get(
+            "/auth/me", headers={"Authorization": f"Bearer {expired_token}"}
+        )
+        assert response.status_code == 401
+
+    def test_refresh_token_as_access_returns_401(self, client, mock_db):
+        """Using a refresh token as an access token should return 401."""
+        user = _make_regular_user()
+        refresh_token = create_refresh_token(user["id"], user["email"], user["role"])
+
+        response = client.get(
+            "/auth/me", headers={"Authorization": f"Bearer {refresh_token}"}
+        )
+        assert response.status_code == 401
+
+
+class TestRefreshToken:
+    """POST /auth/refresh"""
+
+    def test_valid_refresh_token_returns_new_tokens(self, client, mock_db):
+        """A valid refresh token should issue new access and refresh tokens."""
+        user = _make_regular_user()
+        mock_db.get_user_by_id.return_value = user
+        refresh = create_refresh_token(user["id"], user["email"], user["role"])
+
+        response = client.post(
+            "/auth/refresh", json={"refresh_token": refresh}
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert "access_token" in data
+        assert "refresh_token" in data
+
+    def test_invalid_refresh_token_returns_401(self, client, mock_db):
+        """An invalid refresh token should return 401."""
+        response = client.post(
+            "/auth/refresh", json={"refresh_token": "invalid-token-string"}
+        )
+        assert response.status_code == 401
+
+    def test_access_token_as_refresh_returns_401(self, client, mock_db):
+        """Using an access token as a refresh token should return 401."""
+        user = _make_regular_user()
+        access = create_access_token(user["id"], user["email"], user["role"])
+
+        response = client.post(
+            "/auth/refresh", json={"refresh_token": access}
+        )
+        assert response.status_code == 401
+
+
+class TestAdminUsers:
+    """GET /admin/users and PATCH /admin/users/{id}/role"""
+
+    def test_admin_can_list_users(self, client, mock_db):
+        """Admin token should allow listing users."""
+        admin = _make_admin_user()
+        mock_db.get_user_by_id.return_value = admin
+        mock_db.get_all_users.return_value = [admin, _make_regular_user()]
+
+        response = client.get("/admin/users", headers=_auth_header(admin))
+
+        assert response.status_code == 200
+        data = response.json()
+        assert len(data) == 2
+
+    def test_regular_user_cannot_list_users(self, client, mock_db):
+        """Regular user token should be rejected with 403."""
+        user = _make_regular_user()
+        mock_db.get_user_by_id.return_value = user
+
+        response = client.get("/admin/users", headers=_auth_header(user))
+
+        assert response.status_code == 403
+
+    def test_no_token_cannot_list_users(self, client):
+        """No token should be rejected."""
+        response = client.get("/admin/users")
+        assert response.status_code in (401, 403)
+
+    def test_admin_can_change_user_role(self, client, mock_db):
+        """Admin should be able to change another user's role."""
+        admin = _make_admin_user()
+        mock_db.get_user_by_id.return_value = admin
+        mock_db.update_user_role.return_value = {
+            "id": 2,
+            "email": "user@test.com",
+            "role": "admin",
+            "created_at": datetime(2025, 1, 1, tzinfo=timezone.utc),
+        }
+
+        response = client.patch(
+            "/admin/users/2/role",
+            json={"role": "admin"},
+            headers=_auth_header(admin),
+        )
+
+        assert response.status_code == 200
+        assert response.json()["role"] == "admin"
+
+    def test_admin_cannot_change_own_role(self, client, mock_db):
+        """Admin should not be able to change their own role."""
+        admin = _make_admin_user()
+        mock_db.get_user_by_id.return_value = admin
+
+        response = client.patch(
+            "/admin/users/1/role",
+            json={"role": "user"},
+            headers=_auth_header(admin),
+        )
+
+        assert response.status_code == 400
+        assert "own role" in response.json()["detail"].lower()

tests/test_llm.py

@@ -1,72 +1,118 @@
 """Tests for LLM analysis functionality."""
 
+from unittest.mock import Mock
+
 import pytest
-from unittest.mock import Mock, MagicMock
+
 from SPARC.llm import LLMAnalyzer
 
 
 class TestLLMAnalyzer:
     """Test LLM analyzer initialization and API interaction."""
 
+    @pytest.fixture(autouse=True)
+    def mock_database(self, mocker):
+        """Mock the database client for all tests."""
+        mock_db_client = Mock()
+        mock_db_client.get_cached_response.return_value = None  # No cache hit by default
+        mock_db_client.store_message.return_value = 1
+        mocker.patch("SPARC.llm.DatabaseClient", return_value=mock_db_client)
+        return mock_db_client
+
     def test_analyzer_initialization_with_api_key(self, mocker):
         """Test that analyzer initializes with provided API key."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
 
         analyzer = LLMAnalyzer(api_key="test-key-123")
 
-        mock_anthropic.assert_called_once_with(api_key="test-key-123")
-        assert analyzer.model == "claude-3-5-sonnet-20241022"
+        mock_openai.assert_called_once_with(
+            api_key="test-key-123",
+            base_url="https://openrouter.ai/api/v1"
+        )
+        assert analyzer.model == "anthropic/claude-3.5-sonnet"
 
     def test_analyzer_initialization_from_config(self, mocker):
         """Test that analyzer loads API key from config when not provided."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
         mock_config = mocker.patch("SPARC.llm.config")
-        mock_config.anthropic_api_key = "config-key-456"
+        mock_config.openrouter_api_key = "config-key-456"
+        mock_config.use_cache = True
+        mock_config.database_url = "postgresql://localhost/test"
 
         analyzer = LLMAnalyzer()
 
-        mock_anthropic.assert_called_once_with(api_key="config-key-456")
+        mock_openai.assert_called_once_with(
+            api_key="config-key-456",
+            base_url="https://openrouter.ai/api/v1"
+        )
 
-    def test_analyze_patent_content(self, mocker):
+    def test_analyze_patent_content(self, mocker, mock_database):
         """Test single patent content analysis."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
         mock_client = Mock()
-        mock_anthropic.return_value = mock_client
+        mock_openai.return_value = mock_client
 
         # Mock the API response
         mock_response = Mock()
-        mock_response.content = [Mock(text="Innovative GPU architecture.")]
-        mock_client.messages.create.return_value = mock_response
+        mock_response.choices = [Mock(message=Mock(content="Innovative GPU architecture."))]
+        mock_response.usage = Mock(prompt_tokens=100, completion_tokens=50, total_tokens=150)
+        mock_client.chat.completions.create.return_value = mock_response
 
-        analyzer = LLMAnalyzer(api_key="test-key")
+        analyzer = LLMAnalyzer(api_key="test-key", use_cache=False)
         result = analyzer.analyze_patent_content(
             patent_content="ABSTRACT: GPU with new cache design...",
             company_name="NVIDIA",
         )
 
         assert result == "Innovative GPU architecture."
-        mock_client.messages.create.assert_called_once()
+        mock_client.chat.completions.create.assert_called_once()
 
         # Verify the prompt includes company name and content
-        call_args = mock_client.messages.create.call_args
+        call_args = mock_client.chat.completions.create.call_args
         prompt_text = call_args[1]["messages"][0]["content"]
         assert "NVIDIA" in prompt_text
         assert "GPU with new cache design" in prompt_text
 
-    def test_analyze_patent_portfolio(self, mocker):
-        """Test portfolio analysis with multiple patents."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        # Verify message was stored in database
+        mock_database.store_message.assert_called_once()
+
+    def test_analyze_patent_content_cache_hit(self, mocker, mock_database):
+        """Test that cached responses are returned without API call."""
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
         mock_client = Mock()
-        mock_anthropic.return_value = mock_client
+        mock_openai.return_value = mock_client
+
+        # Set up cache hit
+        mock_database.get_cached_response.return_value = {
+            "id": 1,
+            "response": "Cached analysis result"
+        }
+
+        analyzer = LLMAnalyzer(api_key="test-key", use_cache=True)
+        result = analyzer.analyze_patent_content(
+            patent_content="ABSTRACT: GPU with new cache design...",
+            company_name="NVIDIA",
+        )
+
+        assert result == "Cached analysis result"
+        # API should NOT be called on cache hit
+        mock_client.chat.completions.create.assert_not_called()
+
+    def test_analyze_patent_portfolio(self, mocker, mock_database):
+        """Test portfolio analysis with multiple patents."""
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
+        mock_client = Mock()
+        mock_openai.return_value = mock_client
 
         # Mock the API response
         mock_response = Mock()
-        mock_response.content = [
-            Mock(text="Strong portfolio in AI and graphics.")
+        mock_response.choices = [
+            Mock(message=Mock(content="Strong portfolio in AI and graphics."))
         ]
-        mock_client.messages.create.return_value = mock_response
+        mock_response.usage = Mock(prompt_tokens=200, completion_tokens=100, total_tokens=300)
+        mock_client.chat.completions.create.return_value = mock_response
 
-        analyzer = LLMAnalyzer(api_key="test-key")
+        analyzer = LLMAnalyzer(api_key="test-key", use_cache=False)
         patents_data = [
             {"patent_id": "US123", "content": "AI acceleration patent"},
             {"patent_id": "US456", "content": "Graphics rendering patent"},
@@ -77,48 +123,68 @@ class TestLLMAnalyzer:
         )
 
         assert result == "Strong portfolio in AI and graphics."
-        mock_client.messages.create.assert_called_once()
+        mock_client.chat.completions.create.assert_called_once()
 
         # Verify the prompt includes all patents
-        call_args = mock_client.messages.create.call_args
+        call_args = mock_client.chat.completions.create.call_args
         prompt_text = call_args[1]["messages"][0]["content"]
         assert "US123" in prompt_text
         assert "US456" in prompt_text
         assert "AI acceleration patent" in prompt_text
         assert "Graphics rendering patent" in prompt_text
 
-    def test_analyze_patent_portfolio_with_correct_token_limit(self, mocker):
+    def test_analyze_patent_portfolio_with_correct_token_limit(self, mocker, mock_database):
         """Test that portfolio analysis uses higher token limit."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
         mock_client = Mock()
-        mock_anthropic.return_value = mock_client
+        mock_openai.return_value = mock_client
 
         mock_response = Mock()
-        mock_response.content = [Mock(text="Analysis result.")]
-        mock_client.messages.create.return_value = mock_response
+        mock_response.choices = [Mock(message=Mock(content="Analysis result."))]
+        mock_response.usage = Mock(prompt_tokens=100, completion_tokens=50, total_tokens=150)
+        mock_client.chat.completions.create.return_value = mock_response
 
-        analyzer = LLMAnalyzer(api_key="test-key")
+        analyzer = LLMAnalyzer(api_key="test-key", use_cache=False)
         patents_data = [{"patent_id": "US123", "content": "Test content"}]
 
         analyzer.analyze_patent_portfolio(patents_data, "TestCo")
 
-        call_args = mock_client.messages.create.call_args
+        call_args = mock_client.chat.completions.create.call_args
         # Portfolio analysis should use 2048 tokens
         assert call_args[1]["max_tokens"] == 2048
 
-    def test_analyze_single_patent_with_correct_token_limit(self, mocker):
+    def test_analyze_single_patent_with_correct_token_limit(self, mocker, mock_database):
         """Test that single patent analysis uses lower token limit."""
-        mock_anthropic = mocker.patch("SPARC.llm.Anthropic")
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
         mock_client = Mock()
-        mock_anthropic.return_value = mock_client
+        mock_openai.return_value = mock_client
 
         mock_response = Mock()
-        mock_response.content = [Mock(text="Analysis result.")]
-        mock_client.messages.create.return_value = mock_response
+        mock_response.choices = [Mock(message=Mock(content="Analysis result."))]
+        mock_response.usage = Mock(prompt_tokens=100, completion_tokens=50, total_tokens=150)
+        mock_client.chat.completions.create.return_value = mock_response
 
-        analyzer = LLMAnalyzer(api_key="test-key")
+        analyzer = LLMAnalyzer(api_key="test-key", use_cache=False)
         analyzer.analyze_patent_content("Test content", "TestCo")
 
-        call_args = mock_client.messages.create.call_args
+        call_args = mock_client.chat.completions.create.call_args
         # Single patent should use 1024 tokens
         assert call_args[1]["max_tokens"] == 1024
+
+    def test_database_always_initialized(self, mocker, mock_database):
+        """Test that database client is always initialized."""
+        mock_openai = mocker.patch("SPARC.llm.OpenAI")
+
+        analyzer = LLMAnalyzer(api_key="test-key")
+
+        assert analyzer.db_client is not None
+
+    def test_no_api_key_stores_placeholder(self, mocker, mock_database):
+        """Test that without API key, a placeholder is stored."""
+        mocker.patch("SPARC.llm.config")
+
+        analyzer = LLMAnalyzer(use_cache=False)
+        result = analyzer.analyze_patent_content("Test content", "TestCo")
+
+        assert "[NO API]" in result
+        mock_database.store_message.assert_called_once()

tests/test_rate_limit.py

@@ -0,0 +1,97 @@
+"""Tests for rate limiting on auth endpoints."""
+
+import pytest
+from unittest.mock import Mock, patch, MagicMock
+from fastapi.testclient import TestClient
+
+from SPARC.api import app
+
+
+@pytest.fixture
+def client():
+    """Create test client with rate limiter enabled."""
+    return TestClient(app)
+
+
+@pytest.fixture(autouse=True)
+def reset_limiter():
+    """Reset rate limiter storage between tests."""
+    from SPARC.api import limiter
+    limiter.reset()
+    yield
+
+
+class TestRateLimiting:
+    """Test rate limiting on login and register endpoints."""
+
+    @patch("SPARC.api.get_db_client")
+    def test_login_allows_requests_under_limit(self, mock_db_client, client):
+        """Login endpoint allows requests under the rate limit."""
+        mock_db = MagicMock()
+        mock_db.authenticate_user.return_value = None
+        mock_db_client.return_value = mock_db
+
+        # Should allow at least a few requests
+        for _ in range(5):
+            response = client.post(
+                "/auth/login",
+                json={"email": "test@example.com", "password": "password123"},
+            )
+            # 401 is expected (invalid credentials), not 429
+            assert response.status_code == 401
+
+    @patch("SPARC.api.get_db_client")
+    def test_login_rate_limited_after_threshold(self, mock_db_client, client):
+        """Login endpoint returns 429 after exceeding rate limit."""
+        mock_db = MagicMock()
+        mock_db.authenticate_user.return_value = None
+        mock_db_client.return_value = mock_db
+
+        # Send more than the limit (10/minute)
+        statuses = []
+        for _ in range(15):
+            response = client.post(
+                "/auth/login",
+                json={"email": "test@example.com", "password": "password123"},
+            )
+            statuses.append(response.status_code)
+
+        # At least one should be 429
+        assert 429 in statuses, f"Expected 429 in statuses but got: {set(statuses)}"
+
+    @patch("SPARC.api.get_db_client")
+    def test_register_rate_limited_after_threshold(self, mock_db_client, client):
+        """Register endpoint returns 429 after exceeding rate limit."""
+        mock_db = MagicMock()
+        mock_db.get_user_count.return_value = 1
+        mock_db.create_user.return_value = None  # triggers 400 (email exists)
+        mock_db_client.return_value = mock_db
+
+        # Send more than the limit (5/minute)
+        statuses = []
+        for _ in range(10):
+            response = client.post(
+                "/auth/register",
+                json={"email": "test@example.com", "password": "password123"},
+            )
+            statuses.append(response.status_code)
+
+        # At least one should be 429
+        assert 429 in statuses, f"Expected 429 in statuses but got: {set(statuses)}"
+
+    @patch("SPARC.api.get_db_client")
+    def test_rate_limit_returns_retry_after_header(self, mock_db_client, client):
+        """Rate limited responses include a Retry-After header."""
+        mock_db = MagicMock()
+        mock_db.authenticate_user.return_value = None
+        mock_db_client.return_value = mock_db
+
+        # Exhaust the limit
+        for _ in range(15):
+            response = client.post(
+                "/auth/login",
+                json={"email": "test@example.com", "password": "password123"},
+            )
+            if response.status_code == 429:
+                assert "Retry-After" in response.headers
+                break

tests/test_security.py

@@ -0,0 +1,116 @@
+"""Tests for security hardening: JWT secret startup check, CORS config, credential handling."""
+
+import os
+from unittest.mock import patch
+
+import pytest
+
+
+class TestJWTSecretStartupCheck:
+    """Test the startup guard that refuses default JWT secret in non-dev environments."""
+
+    def test_default_secret_in_production_raises(self):
+        """Starting with default secret and APP_ENV=production must raise RuntimeError."""
+        with patch.dict(os.environ, {"APP_ENV": "production"}):
+            # Reload config to pick up the new APP_ENV
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+
+            from SPARC.auth import _DEFAULT_JWT_SECRET, check_jwt_secret
+            # Patch JWT_SECRET to the default
+            with patch("SPARC.auth.JWT_SECRET", _DEFAULT_JWT_SECRET):
+                with pytest.raises(RuntimeError, match="FATAL.*JWT_SECRET"):
+                    check_jwt_secret()
+
+            # Restore config
+            with patch.dict(os.environ, {"APP_ENV": "development"}):
+                importlib.reload(SPARC.config)
+
+    def test_default_secret_in_development_succeeds(self):
+        """Starting with default secret and APP_ENV=development must not raise."""
+        with patch.dict(os.environ, {"APP_ENV": "development"}):
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+
+            from SPARC.auth import _DEFAULT_JWT_SECRET, check_jwt_secret
+            with patch("SPARC.auth.JWT_SECRET", _DEFAULT_JWT_SECRET):
+                # Should not raise
+                check_jwt_secret()
+
+            # Restore
+            importlib.reload(SPARC.config)
+
+    def test_custom_secret_in_production_succeeds(self):
+        """Starting with a custom secret in production must not raise."""
+        with patch.dict(os.environ, {"APP_ENV": "production"}):
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+
+            from SPARC.auth import check_jwt_secret
+            with patch("SPARC.auth.JWT_SECRET", "my-secure-random-secret-abc123"):
+                # Should not raise
+                check_jwt_secret()
+
+            with patch.dict(os.environ, {"APP_ENV": "development"}):
+                importlib.reload(SPARC.config)
+
+    def test_default_secret_unset_env_succeeds(self):
+        """When APP_ENV is unset (defaults to development), default secret is allowed."""
+        with patch.dict(os.environ, {}, clear=False):
+            # Remove APP_ENV if present
+            env = os.environ.copy()
+            env.pop("APP_ENV", None)
+            with patch.dict(os.environ, env, clear=True):
+                import importlib
+                import SPARC.config
+                importlib.reload(SPARC.config)
+
+                from SPARC.auth import _DEFAULT_JWT_SECRET, check_jwt_secret
+                with patch("SPARC.auth.JWT_SECRET", _DEFAULT_JWT_SECRET):
+                    # Should not raise (defaults to development)
+                    check_jwt_secret()
+
+                with patch.dict(os.environ, {"APP_ENV": "development"}):
+                    importlib.reload(SPARC.config)
+
+
+class TestCORSConfig:
+    """Test that CORS origins are configurable via environment variable."""
+
+    def test_default_cors_origins(self):
+        """When CORS_ORIGINS is unset, defaults to localhost origins."""
+        with patch.dict(os.environ, {"CORS_ORIGINS": ""}):
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+            assert SPARC.config.cors_origins == [
+                "http://localhost:3000",
+                "http://localhost:5173",
+            ]
+
+    def test_custom_cors_origins(self):
+        """Setting CORS_ORIGINS configures allowed origins."""
+        with patch.dict(os.environ, {"CORS_ORIGINS": "https://sparc.example.com,https://app.example.com"}):
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+            assert SPARC.config.cors_origins == [
+                "https://sparc.example.com",
+                "https://app.example.com",
+            ]
+            # Restore
+            with patch.dict(os.environ, {"CORS_ORIGINS": ""}):
+                importlib.reload(SPARC.config)
+
+    def test_single_cors_origin(self):
+        """A single origin without comma works correctly."""
+        with patch.dict(os.environ, {"CORS_ORIGINS": "https://sparc.example.com"}):
+            import importlib
+            import SPARC.config
+            importlib.reload(SPARC.config)
+            assert SPARC.config.cors_origins == ["https://sparc.example.com"]
+            with patch.dict(os.environ, {"CORS_ORIGINS": ""}):
+                importlib.reload(SPARC.config)

tests/test_serp_api.py

@@ -1,7 +1,10 @@
 """Tests for SERP API patent retrieval and parsing functionality."""
 
-import pytest
+from datetime import datetime, timedelta
+from unittest.mock import Mock
+
 from SPARC.serp_api import SERP
+from SPARC.types import Patent
 
 
 class TestTextCleaning:
@@ -176,3 +179,89 @@ class TestPatentMinimization:
 
         # Sections should be separated by double newlines
         assert "\n\n" in result
+
+
+class TestDynamicDateRange:
+    """Test dynamic date range computation in SERP.query."""
+
+    def test_query_uses_rolling_date_window(self, mocker):
+        """Verify the date filter uses a rolling window, not hardcoded dates."""
+        mock_search = mocker.patch("SPARC.serp_api.serpapi.search")
+        mock_search.return_value = {"organic_results": []}
+        mocker.patch("SPARC.serp_api.config.api_key", "fake-key")
+        mocker.patch("SPARC.serp_api.config.patent_search_days", 90)
+
+        SERP.query("TestCorp")
+
+        call_params = mock_search.call_args[0][0]
+        tbs = call_params["tbs"]
+        # Should contain "cdr:1,cd_min:" with a date, not the old hardcoded one
+        assert "cdr:1,cd_min:" in tbs
+        assert "10/28/2025" not in tbs  # old hardcoded date gone
+
+    def test_query_respects_days_back_param(self, mocker):
+        """Verify days_back parameter controls the date window."""
+        mock_search = mocker.patch("SPARC.serp_api.serpapi.search")
+        mock_search.return_value = {"organic_results": []}
+        mocker.patch("SPARC.serp_api.config.api_key", "fake-key")
+        mocker.patch("SPARC.serp_api.config.patent_search_days", 90)
+
+        now = datetime.now()
+        SERP.query("TestCorp", days_back=30)
+
+        call_params = mock_search.call_args[0][0]
+        tbs = call_params["tbs"]
+        expected_start = (now - timedelta(days=30)).strftime("%-m/%-d/%Y")
+        assert expected_start in tbs
+
+
+class TestFilesystemPDFCaching:
+    """Test that save_patents skips download for existing files."""
+
+    def test_save_patents_skips_download_when_cached(self, mocker, tmp_path):
+        """Already-downloaded PDFs should not be re-downloaded."""
+        mock_get = mocker.patch("SPARC.serp_api.requests.get")
+        mocker.patch("SPARC.serp_api.os.makedirs")
+
+        pdf_path = tmp_path / "US123.pdf"
+        pdf_path.write_bytes(b"%PDF-1.4 fake content")
+
+        mocker.patch("SPARC.serp_api.os.path.exists", return_value=True)
+        mocker.patch("SPARC.serp_api.os.path.getsize", return_value=100)
+
+        patent = Patent(patent_id="US123", pdf_link="http://example.com/test.pdf")
+        result = SERP.save_patents(patent)
+
+        mock_get.assert_not_called()
+        assert result.pdf_path == "patents/US123.pdf"
+
+    def test_save_patents_downloads_when_not_cached(self, mocker):
+        """Missing PDFs should be downloaded."""
+        mock_response = Mock()
+        mock_response.content = b"%PDF-1.4 content"
+        mock_get = mocker.patch("SPARC.serp_api.requests.get", return_value=mock_response)
+        mocker.patch("SPARC.serp_api.os.makedirs")
+        mocker.patch("SPARC.serp_api.os.path.exists", return_value=False)
+        mock_open = mocker.patch("builtins.open", mocker.mock_open())
+
+        patent = Patent(patent_id="US456", pdf_link="http://example.com/test.pdf")
+        result = SERP.save_patents(patent)
+
+        mock_get.assert_called_once_with("http://example.com/test.pdf")
+        assert result.pdf_path == "patents/US456.pdf"
+
+    def test_save_patents_redownloads_empty_files(self, mocker):
+        """Empty/corrupt PDFs (0 bytes) should be re-downloaded."""
+        mock_response = Mock()
+        mock_response.content = b"%PDF-1.4 content"
+        mock_get = mocker.patch("SPARC.serp_api.requests.get", return_value=mock_response)
+        mocker.patch("SPARC.serp_api.os.makedirs")
+        mocker.patch("SPARC.serp_api.os.path.exists", return_value=True)
+        mocker.patch("SPARC.serp_api.os.path.getsize", return_value=0)
+        mock_open = mocker.patch("builtins.open", mocker.mock_open())
+
+        patent = Patent(patent_id="US789", pdf_link="http://example.com/test.pdf")
+        result = SERP.save_patents(patent)
+
+        mock_get.assert_called_once()
+        assert result.pdf_path == "patents/US789.pdf"