Docs: add MODEL, SERP_CACHE_TTL_HOURS, and LOG_LEVEL to .env.example

These environment variables were already supported in config.py but
were not documented in .env.example, making them hard to discover.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.env.example

@@ -35,8 +35,41 @@ JWT_SECRET=your-secure-jwt-secret-change-in-production
 # 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/test.yaml

@@ -34,6 +34,17 @@ jobs:
         run: |
           ruff check SPARC/ tests/
 
+      - name: Install Node.js and frontend dependencies
+        shell: sh
+        run: |
+          apk add --no-cache nodejs npm
+          cd frontend && npm ci
+
+      - name: Run TypeScript type check
+        shell: sh
+        run: |
+          cd frontend && npx tsc --noEmit
+
       - name: Run pytest
         shell: sh
         env:

SPARC/analyzer.py

@@ -108,12 +108,10 @@ class CompanyAnalyzer:
     def analyze_single_patent(self, patent_id: str, company_name: str) -> str:
         """Analyze a single patent by ID.
 
-        Prerequisite:
-            The patent PDF must already exist at ``patents/{patent_id}.pdf``
-            before calling this method. PDFs are downloaded automatically when
-            using the batch analysis pipeline (``analyze_company`` or the
-            ``/analyze/batch`` API endpoint). For standalone usage, download
-            the PDF manually or call ``SERP.save_patents()`` first.
+        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")
@@ -123,7 +121,7 @@ class CompanyAnalyzer:
           Analysis of the specific patent's innovation quality
 
         Raises:
-          FileNotFoundError: If the patent PDF is not found at the expected path.
+          FileNotFoundError: If the patent PDF cannot be found or downloaded.
         """
         import os
         logger.info("Analyzing patent %s for %s...", patent_id, company_name)
@@ -131,10 +129,22 @@ class CompanyAnalyzer:
         patent_path = f"patents/{patent_id}.pdf"
 
         if not os.path.exists(patent_path):
-            raise FileNotFoundError(
-                f"Patent PDF not found at '{patent_path}'. "
-                f"Download the PDF first using SERP.save_patents() or the batch analysis pipeline."
-            )
+            # 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)

SPARC/api.py

@@ -9,7 +9,7 @@ from typing import Annotated, List
 
 from fastapi import BackgroundTasks, Depends, FastAPI, HTTPException, Query, Request
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import JSONResponse
+from fastapi.responses import JSONResponse, StreamingResponse
 from pydantic import BaseModel, EmailStr, Field
 from slowapi import Limiter
 from slowapi.errors import RateLimitExceeded
@@ -41,6 +41,7 @@ class CompanyAnalysisResponse(BaseModel):
     patent_count: int
     success: bool
     error: str | None = None
+    model: str | None = None
     timestamp: datetime
 
 
@@ -54,6 +55,15 @@ class BatchAnalysisResponse(BaseModel):
     timestamp: datetime
 
 
+class CompanyAnalysisRequest(BaseModel):
+    """Request model for single company analysis with optional model selection."""
+
+    model: str | None = Field(
+        default=None,
+        description="LLM model to use (e.g. 'anthropic/claude-3.5-sonnet', 'openai/gpt-4o'). Defaults to server config.",
+    )
+
+
 class BatchAnalysisRequest(BaseModel):
     """Request model for batch company analysis."""
 
@@ -63,6 +73,10 @@ class BatchAnalysisRequest(BaseModel):
     max_workers: int = Field(
         default=3, ge=1, le=5, description="Max concurrent analyses"
     )
+    model: str | None = Field(
+        default=None,
+        description="LLM model to use for all analyses in this batch. Defaults to server config.",
+    )
 
 
 class JobStatus(BaseModel):
@@ -77,6 +91,13 @@ class JobStatus(BaseModel):
     error: str | None = None
 
 
+class PaginatedJobsResponse(BaseModel):
+    """Paginated response for job listings."""
+
+    items: list["JobStatus"]
+    next_cursor: str | None = None
+
+
 class HealthResponse(BaseModel):
     """Health check response."""
 
@@ -133,6 +154,7 @@ def _convert_result(result: CompanyAnalysisResult) -> CompanyAnalysisResponse:
         patent_count=result.patent_count,
         success=result.success,
         error=result.error,
+        model=result.model,
         timestamp=result.timestamp,
     )
 
@@ -169,6 +191,9 @@ async def lifespan(app: FastAPI):
         import logging
         logging.getLogger(__name__).warning("Marked %d stale jobs as failed on startup", stale)
     _db.close()
+    # Start scheduled analysis if tracked companies are configured
+    from SPARC.scheduler import start_scheduler
+    start_scheduler()
     yield
     # Cleanup
     _analyzer = None
@@ -369,6 +394,60 @@ async def delete_user(
     return {"message": "User deleted"}
 
 
+# ============== Tracked Companies Endpoints ==============
+
+
+class TrackCompanyRequest(BaseModel):
+    """Request to add a company to tracking."""
+
+    company_name: str = Field(..., min_length=1, max_length=255)
+
+
+@app.get("/admin/tracked", tags=["Admin"])
+async def list_tracked_companies(
+    _: UserResponse = Depends(get_current_admin),
+):
+    """List all tracked companies (admin only)."""
+    db = get_db_client()
+    return db.list_tracked_companies()
+
+
+@app.post("/admin/tracked", tags=["Admin"])
+async def add_tracked_company(
+    request: TrackCompanyRequest,
+    _: UserResponse = Depends(get_current_admin),
+):
+    """Add a company to the tracked list (admin only)."""
+    db = get_db_client()
+    result = db.add_tracked_company(request.company_name)
+    if not result:
+        raise HTTPException(status_code=409, detail="Company already tracked")
+    return result
+
+
+@app.delete("/admin/tracked/{company_name}", tags=["Admin"])
+async def remove_tracked_company(
+    company_name: str,
+    _: UserResponse = Depends(get_current_admin),
+):
+    """Remove a company from the tracked list (admin only)."""
+    db = get_db_client()
+    removed = db.remove_tracked_company(company_name)
+    if not removed:
+        raise HTTPException(status_code=404, detail="Company not found in tracking list")
+    return {"message": f"Stopped tracking {company_name}"}
+
+
+@app.get("/admin/alerts", tags=["Admin"])
+async def list_alerts(
+    limit: int = Query(default=50, ge=1, le=200),
+    _: UserResponse = Depends(get_current_admin),
+):
+    """List recent alerts from scheduled analysis (admin only)."""
+    db = get_db_client()
+    return db.list_alerts(limit=limit)
+
+
 # ============== Analytics Endpoint ==============
 
 
@@ -389,6 +468,317 @@ async def get_analytics(
     )
 
 
+# ============== Model Selection Endpoints ==============
+
+# Supported models via OpenRouter
+SUPPORTED_MODELS = [
+    {"id": "anthropic/claude-3.5-sonnet", "name": "Claude 3.5 Sonnet", "provider": "Anthropic"},
+    {"id": "openai/gpt-4o", "name": "GPT-4o", "provider": "OpenAI"},
+    {"id": "openai/gpt-4o-mini", "name": "GPT-4o Mini", "provider": "OpenAI"},
+    {"id": "google/gemini-pro-1.5", "name": "Gemini Pro 1.5", "provider": "Google"},
+    {"id": "meta-llama/llama-3.1-70b-instruct", "name": "Llama 3.1 70B", "provider": "Meta"},
+]
+
+
+@app.get("/models", tags=["System"])
+async def list_models():
+    """List supported LLM models for analysis.
+
+    Returns the available models that can be passed as the `model` field
+    in analysis requests. The default model is determined by the `MODEL`
+    environment variable on the server.
+    """
+    return {
+        "models": SUPPORTED_MODELS,
+        "default": config.model,
+    }
+
+
+@app.get("/analytics/trends", tags=["Analytics"])
+async def get_analytics_trends(
+    days: int = Query(default=90, ge=7, le=365),
+    _: UserResponse = Depends(get_current_user),
+):
+    """Get trend data for patent analysis over time.
+
+    Returns two datasets:
+    - ``by_month``: analysis count per company per month
+    - ``by_type_over_time``: analysis type distribution per month
+
+    Args:
+        days: Number of days to look back (default 90)
+
+    Returns:
+        Trend data suitable for time-series and distribution charts
+    """
+    db = get_db_client()
+
+    with db.get_conn() as conn:
+        with conn.cursor() as cur:
+            # Analyses per company per month
+            cur.execute(
+                """
+                SELECT
+                    TO_CHAR(timestamp, 'YYYY-MM') AS month,
+                    company_name,
+                    COUNT(*) AS count
+                FROM llm_messages
+                WHERE timestamp >= NOW() - INTERVAL '%s days'
+                    AND is_cached = FALSE
+                    AND company_name IS NOT NULL
+                GROUP BY month, company_name
+                ORDER BY month
+                """,
+                (days,),
+            )
+            by_month_rows = cur.fetchall()
+
+            # Analysis type distribution per month
+            cur.execute(
+                """
+                SELECT
+                    TO_CHAR(timestamp, 'YYYY-MM') AS month,
+                    analysis_type,
+                    COUNT(*) AS count
+                FROM llm_messages
+                WHERE timestamp >= NOW() - INTERVAL '%s days'
+                    AND is_cached = FALSE
+                GROUP BY month, analysis_type
+                ORDER BY month
+                """,
+                (days,),
+            )
+            by_type_rows = cur.fetchall()
+
+    by_month = [
+        {"month": row[0], "company_name": row[1], "count": row[2]}
+        for row in by_month_rows
+    ]
+    by_type_over_time = [
+        {"month": row[0], "analysis_type": row[1], "count": row[2]}
+        for row in by_type_rows
+    ]
+
+    return {
+        "by_month": by_month,
+        "by_type_over_time": by_type_over_time,
+        "period_days": days,
+    }
+
+
+# ============== Export Endpoints ==============
+
+
+@app.get("/export/{company_name}", tags=["Export"])
+async def export_company_csv(
+    company_name: str,
+    _: UserResponse = Depends(get_current_user),
+):
+    """Export analysis results for a company as a CSV file.
+
+    Returns all stored analysis records for the given company, including
+    analysis type, model used, response text, and timestamp.
+
+    Args:
+        company_name: Company name to export results for
+
+    Returns:
+        CSV file download
+    """
+    import csv
+    import io
+
+    db = get_db_client()
+    # Query all non-cached analysis results for this company
+    with db.get_conn() as conn:
+        with conn.cursor() as cur:
+            cur.execute(
+                """
+                SELECT company_name, analysis_type, model, response, timestamp
+                FROM llm_messages
+                WHERE LOWER(company_name) = LOWER(%s) AND is_cached = FALSE
+                ORDER BY timestamp DESC
+                """,
+                (company_name,),
+            )
+            rows = cur.fetchall()
+
+    if not rows:
+        raise HTTPException(status_code=404, detail=f"No analysis results found for '{company_name}'")
+
+    output = io.StringIO()
+    writer = csv.writer(output)
+    writer.writerow(["company_name", "analysis_type", "model", "analysis", "timestamp"])
+    for row in rows:
+        writer.writerow(row)
+
+    output.seek(0)
+    safe_name = company_name.replace(" ", "_").lower()
+    return StreamingResponse(
+        iter([output.getvalue()]),
+        media_type="text/csv",
+        headers={"Content-Disposition": f'attachment; filename="sparc_{safe_name}_export.csv"'},
+    )
+
+
+@app.get("/export/{company_name}/pdf", tags=["Export"])
+async def export_company_pdf(
+    company_name: str,
+    _: UserResponse = Depends(get_current_user),
+):
+    """Export analysis results for a company as a formatted PDF report.
+
+    Returns all stored analysis records for the given company, including
+    analysis type, model used, response text, and timestamp, formatted
+    as a downloadable PDF document.
+
+    Args:
+        company_name: Company name to export results for
+
+    Returns:
+        PDF file download
+    """
+    import io
+    import textwrap
+
+    from reportlab.lib import colors
+    from reportlab.lib.pagesizes import letter
+    from reportlab.lib.styles import ParagraphStyle, getSampleStyleSheet
+    from reportlab.lib.units import inch
+    from reportlab.platypus import (
+        Paragraph,
+        SimpleDocTemplate,
+        Spacer,
+        Table,
+        TableStyle,
+    )
+
+    db = get_db_client()
+    with db.get_conn() as conn:
+        with conn.cursor() as cur:
+            cur.execute(
+                """
+                SELECT company_name, analysis_type, model, response, timestamp
+                FROM llm_messages
+                WHERE LOWER(company_name) = LOWER(%s) AND is_cached = FALSE
+                ORDER BY timestamp DESC
+                """,
+                (company_name,),
+            )
+            rows = cur.fetchall()
+
+    if not rows:
+        raise HTTPException(status_code=404, detail=f"No analysis results found for '{company_name}'")
+
+    buffer = io.BytesIO()
+    doc = SimpleDocTemplate(
+        buffer,
+        pagesize=letter,
+        rightMargin=0.75 * inch,
+        leftMargin=0.75 * inch,
+        topMargin=0.75 * inch,
+        bottomMargin=0.75 * inch,
+    )
+
+    styles = getSampleStyleSheet()
+    title_style = ParagraphStyle(
+        "CustomTitle",
+        parent=styles["Title"],
+        fontSize=20,
+        spaceAfter=6,
+    )
+    subtitle_style = ParagraphStyle(
+        "Subtitle",
+        parent=styles["Normal"],
+        fontSize=11,
+        textColor=colors.grey,
+        spaceAfter=20,
+    )
+    heading_style = ParagraphStyle(
+        "SectionHeading",
+        parent=styles["Heading2"],
+        fontSize=13,
+        spaceBefore=16,
+        spaceAfter=8,
+        textColor=colors.HexColor("#1a1a2e"),
+    )
+    body_style = ParagraphStyle(
+        "BodyText",
+        parent=styles["Normal"],
+        fontSize=9,
+        leading=13,
+        spaceAfter=10,
+    )
+
+    elements = []
+
+    # Title and date
+    display_name = rows[0][0]  # Use the casing from the database
+    analysis_date = datetime.now().strftime("%Y-%m-%d")
+    elements.append(Paragraph(f"SPARC Analysis Report: {display_name}", title_style))
+    elements.append(Paragraph(f"Generated on {analysis_date}", subtitle_style))
+
+    # Summary table
+    summary_data = [
+        ["Total Analyses", str(len(rows))],
+        ["Analysis Types", ", ".join(sorted(set(r[1] for r in rows)))],
+        ["Models Used", ", ".join(sorted(set(r[2] for r in rows)))],
+    ]
+    summary_table = Table(summary_data, colWidths=[2 * inch, 4.5 * inch])
+    summary_table.setStyle(
+        TableStyle(
+            [
+                ("BACKGROUND", (0, 0), (0, -1), colors.HexColor("#f0f0f5")),
+                ("FONTNAME", (0, 0), (0, -1), "Helvetica-Bold"),
+                ("FONTSIZE", (0, 0), (-1, -1), 9),
+                ("PADDING", (0, 0), (-1, -1), 6),
+                ("GRID", (0, 0), (-1, -1), 0.5, colors.HexColor("#cccccc")),
+                ("VALIGN", (0, 0), (-1, -1), "TOP"),
+            ]
+        )
+    )
+    elements.append(summary_table)
+    elements.append(Spacer(1, 16))
+
+    # Individual analysis sections
+    for i, row in enumerate(rows, 1):
+        _, analysis_type, model, response, timestamp = row
+        ts_str = timestamp.strftime("%Y-%m-%d %H:%M:%S") if hasattr(timestamp, "strftime") else str(timestamp)
+
+        elements.append(
+            Paragraph(f"Analysis {i}: {analysis_type} (via {model})", heading_style)
+        )
+        elements.append(
+            Paragraph(f"<i>Performed: {ts_str}</i>", body_style)
+        )
+
+        # Wrap long response text into paragraphs, escaping XML special chars
+        safe_response = (
+            response.replace("&", "&amp;")
+            .replace("<", "&lt;")
+            .replace(">", "&gt;")
+        )
+        # Split into manageable paragraphs to avoid overflow
+        for line in safe_response.split("\n"):
+            if line.strip():
+                elements.append(Paragraph(line, body_style))
+            else:
+                elements.append(Spacer(1, 4))
+
+        elements.append(Spacer(1, 10))
+
+    doc.build(elements)
+    buffer.seek(0)
+
+    safe_name = company_name.replace(" ", "_").lower()
+    filename = f"{safe_name}-analysis-{analysis_date}.pdf"
+    return StreamingResponse(
+        iter([buffer.getvalue()]),
+        media_type="application/pdf",
+        headers={"Content-Disposition": f'attachment; filename="{filename}"'},
+    )
+
+
 # ============== System Endpoints ==============
 
 
@@ -429,6 +819,38 @@ async def analyze_company(
     return _convert_result(result)
 
 
+@app.get(
+    "/analyze/patent/{patent_id}",
+    tags=["Analysis"],
+)
+async def analyze_single_patent(
+    patent_id: str,
+    company_name: str = Query(description="Company name for analysis context"),
+    _: UserResponse = Depends(get_current_user),
+):
+    """Analyze a single patent by its publication ID.
+
+    If the patent PDF is not already cached locally, the system will attempt
+    to download it automatically from a previously cached link. If no link
+    is available, a 404 error is returned.
+
+    Args:
+        patent_id: Patent publication ID (e.g. "US-11234567-B2")
+        company_name: Company name for analysis context
+
+    Returns:
+        Analysis text for the patent
+    """
+    if not _analyzer:
+        raise HTTPException(status_code=503, detail="Analyzer not initialized")
+
+    try:
+        analysis = _analyzer.analyze_single_patent(patent_id, company_name)
+        return {"patent_id": patent_id, "company_name": company_name, "analysis": analysis}
+    except FileNotFoundError as e:
+        raise HTTPException(status_code=404, detail=str(e))
+
+
 @app.post(
     "/analyze/batch",
     response_model=BatchAnalysisResponse,
@@ -519,8 +941,25 @@ def _run_batch_job(job_id: str, companies: list[str], max_workers: int):
             progress=100,
             result_json=_json.dumps(batch_response.model_dump(), default=str),
         )
+        # Fire webhook notification
+        from SPARC.webhooks import notify_job_completed
+        notify_job_completed(
+            job_id=job_id,
+            status="completed",
+            total_companies=result.total_companies,
+            successful=result.successful,
+            failed=result.failed,
+        )
     except Exception as e:
         db.update_job(job_id, status="failed", error=str(e))
+        from SPARC.webhooks import notify_job_completed
+        notify_job_completed(
+            job_id=job_id,
+            status="failed",
+            total_companies=len(companies),
+            successful=0,
+            failed=len(companies),
+        )
 
 
 @app.post("/analyze/batch/async", response_model=JobStatus, tags=["Analysis"])
@@ -577,24 +1016,51 @@ async def get_job_status(
     return _job_row_to_status(job_row)
 
 
-@app.get("/jobs", response_model=list[JobStatus], tags=["Jobs"])
+@app.get("/jobs", response_model=PaginatedJobsResponse, tags=["Jobs"])
 async def list_jobs(
     status: Annotated[
         str | None,
         Query(description="Filter by status: pending, running, completed, failed"),
     ] = None,
     limit: Annotated[int, Query(ge=1, le=100)] = 10,
+    cursor: Annotated[
+        str | None,
+        Query(description="Opaque cursor from a previous response's next_cursor field"),
+    ] = None,
     _: UserResponse = Depends(get_current_user),
 ):
-    """List all analysis jobs.
+    """List analysis jobs with cursor-based pagination.
+
+    Pass ``limit`` to control page size. The response includes a ``next_cursor``
+    field; pass it back as the ``cursor`` query parameter to fetch the next page.
+    When ``next_cursor`` is ``null``, there are no more results.
+
+    Existing clients that use only ``limit`` (without ``cursor``) continue to
+    work without modification.
 
     Args:
         status: Optional filter by job status
         limit: Maximum number of jobs to return (default 10, max 100)
+        cursor: Opaque pagination cursor from a previous response
 
     Returns:
-        List of job statuses
+        Paginated list of job statuses
     """
     db = _get_job_db()
-    job_rows = db.list_jobs(status=status, limit=limit)
-    return [_job_row_to_status(row) for row in job_rows]
+    # Fetch one extra to determine if there is a next page
+    job_rows = db.list_jobs(status=status, limit=limit + 1, cursor=cursor)
+
+    has_next = len(job_rows) > limit
+    if has_next:
+        job_rows = job_rows[:limit]
+
+    items = [_job_row_to_status(row) for row in job_rows]
+
+    next_cursor = None
+    if has_next and job_rows:
+        last = job_rows[-1]
+        created = last["created_at"]
+        ts = created.isoformat() if hasattr(created, "isoformat") else str(created)
+        next_cursor = f"{ts}|{last['job_id']}"
+
+    return PaginatedJobsResponse(items=items, next_cursor=next_cursor)

SPARC/config.py

@@ -53,6 +53,13 @@ root_path = os.getenv("ROOT_PATH", "")
 # 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", "")

SPARC/database.py

@@ -192,6 +192,35 @@ class DatabaseClient:
                 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
@@ -568,20 +597,45 @@ class DatabaseClient:
         self,
         status: Optional[str] = None,
         limit: int = 10,
+        cursor: Optional[str] = None,
     ) -> List[Dict]:
-        """List jobs, optionally filtered by status."""
-        query = "SELECT * FROM jobs"
+        """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:
-            query += " WHERE status = %s"
+            conditions.append("status = %s")
             params.append(status)
-        query += " ORDER BY created_at DESC LIMIT %s"
+
+        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 cursor:
-                cursor.execute(query, params)
-                return [dict(row) for row in cursor.fetchall()]
+            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'.
@@ -803,3 +857,81 @@ class DatabaseClient:
             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

@@ -40,12 +40,13 @@ class LLMAnalyzer:
         else:
             self.client = None
 
-    def analyze_patent_content(self, patent_content: str, company_name: str) -> str:
+    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
@@ -63,6 +64,8 @@ Patent Content:
 
 Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals about the company's technical direction and competitive advantage."""
 
+        effective_model = model or self.model
+
         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]"
@@ -81,7 +84,7 @@ Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals
                     response=cached["response"],
                     company_name=company_name,
                     analysis_type="single_patent",
-                    model=self.model,
+                    model=effective_model,
                     metadata={
                         "patent_content_length": len(patent_content),
                         "cache_hit": True,
@@ -94,7 +97,7 @@ Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals
         # Call API if no cache hit and client is available
         if self.client:
             response = self.client.chat.completions.create(
-                model=self.model,
+                model=effective_model,
                 max_tokens=1024,
                 messages=[{"role": "user", "content": prompt}],
             )
@@ -106,7 +109,7 @@ Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals
                 response=response_text,
                 company_name=company_name,
                 analysis_type="single_patent",
-                model=self.model,
+                model=effective_model,
                 metadata={"patent_content_length": len(patent_content)},
                 token_usage={
                     "prompt_tokens": response.usage.prompt_tokens,
@@ -124,13 +127,13 @@ Provide a concise analysis (2-3 paragraphs) focusing on what this patent reveals
             response=placeholder,
             company_name=company_name,
             analysis_type="single_patent",
-            model=self.model,
+            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.
 
@@ -165,13 +168,16 @@ Patent Portfolio:
 
 Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the company's innovation strength and performance outlook."""
 
+        effective_model = model or self.model
+
         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]
+            "patent_ids": [p['patent_id'] for p in patents_data],
+            "model": effective_model,
         }
 
         # Check cache first
@@ -188,7 +194,7 @@ Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the co
                     response=cached["response"],
                     company_name=company_name,
                     analysis_type="portfolio",
-                    model=self.model,
+                    model=effective_model,
                     metadata={
                         **metadata,
                         "cache_hit": True,
@@ -202,7 +208,7 @@ Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the co
         if self.client:
             try:
                 response = self.client.chat.completions.create(
-                    model=self.model,
+                    model=effective_model,
                     max_tokens=2048,
                     messages=[{"role": "user", "content": prompt}],
                 )
@@ -215,7 +221,7 @@ Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the co
                     response=response_text,
                     company_name=company_name,
                     analysis_type="portfolio",
-                    model=self.model,
+                    model=effective_model,
                     metadata=metadata,
                     token_usage={
                         "prompt_tokens": response.usage.prompt_tokens,
@@ -235,7 +241,7 @@ Provide a comprehensive analysis (4-5 paragraphs) with a final verdict on the co
             response=placeholder,
             company_name=company_name,
             analysis_type="portfolio",
-            model=self.model,
+            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

@@ -1,4 +1,5 @@
-import os
+import io
+import logging
 import re
 from datetime import datetime, timedelta
 from typing import Dict
@@ -8,8 +9,21 @@ import requests
 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, days_back: int = None) -> Patents:
@@ -44,6 +58,7 @@ class SERP:
       "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 results to Patent objects, skipping any without PDF links
     patent_ids = []
@@ -52,13 +67,16 @@ class SERP:
         pdf_link = patent.get("pdf")
         if pdf_link:
             patent_ids.append(Patent(patent_id=patent["publication_number"], pdf_link=pdf_link, summary=None))
-        # Patents without PDF links are skipped (see docstring for details)
+        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, skipping download if already cached.
+    """Save the patent PDF to storage, skipping download if already cached.
+
+    Uses the configured storage backend (local filesystem or S3).
 
     Args:
       patent: Patent object
@@ -66,35 +84,51 @@ class SERP:
     Returns:
       Patent object with updated PDF path
     """
-    pdf_path = f"patents/{patent.patent_id}.pdf"
-    os.makedirs("patents", exist_ok=True)
+    storage = _get_storage()
+    key = f"{patent.patent_id}.pdf"
 
-    if not (os.path.exists(pdf_path) and os.path.getsize(pdf_path) > 0):
+    if not storage.exists(key):
+      logger.info("Downloading PDF for %s", patent.patent_id)
       response = requests.get(patent.pdf_link)
-      with open(pdf_path, "wb") as f:
-        f.write(response.content)
+      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 = pdf_path
+    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

@@ -24,6 +24,7 @@ class CompanyAnalysisResult:
     patent_count: int
     success: bool
     error: str | None = None
+    model: str | None = None
     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

@@ -52,6 +52,29 @@ services:
       - ./patents:/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:
     build: ./frontend
     container_name: sparc-dashboard
@@ -63,3 +86,4 @@ services:
 
 volumes:
   postgres_data:
+  minio_data:

frontend/index.html

@@ -7,6 +7,15 @@
     <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>

frontend/package-lock.json

@@ -10,7 +10,7 @@
       "dependencies": {
         "@tanstack/react-query": "^5.51.0",
         "axios": "^1.7.2",
-        "lucide-react": "^0.400.0",
+        "lucide-react": "^1.7.0",
         "react": "^18.3.1",
         "react-dom": "^18.3.1",
         "react-router-dom": "^6.24.0",
@@ -3452,9 +3452,9 @@
       }
     },
     "node_modules/lucide-react": {
-      "version": "0.400.0",
-      "resolved": "https://registry.npmjs.org/lucide-react/-/lucide-react-0.400.0.tgz",
-      "integrity": "sha512-rpp7pFHh3Xd93KHixNgB0SqThMHpYNzsGUu69UaQbSZ75Q/J3m5t6EhKyMT3m4w2WOxmJ2mY0tD3vebnXqQryQ==",
+      "version": "1.7.0",
+      "resolved": "https://registry.npmjs.org/lucide-react/-/lucide-react-1.7.0.tgz",
+      "integrity": "sha512-yI7BeItCLZJTXikmK4KNUGCKoGzSvbKlfCvw44bU4fXAL6v3gYS4uHD1jzsLkfwODYwI6Drw5Tu9Z5ulDe0TSg==",
       "license": "ISC",
       "peerDependencies": {
         "react": "^16.5.1 || ^17.0.0 || ^18.0.0 || ^19.0.0"

frontend/package.json

@@ -7,12 +7,15 @@
     "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": "^0.400.0",
+    "lucide-react": "^1.7.0",
     "react": "^18.3.1",
     "react-dom": "^18.3.1",
     "react-router-dom": "^6.24.0",
@@ -30,6 +33,7 @@
     "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/src/App.tsx

@@ -1,6 +1,7 @@
 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';
@@ -10,6 +11,7 @@ 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: {
@@ -22,6 +24,7 @@ const queryClient = new QueryClient({
 
 function App() {
   return (
+    <ThemeProvider>
     <QueryClientProvider client={queryClient}>
       <AuthProvider>
         <BrowserRouter>
@@ -41,6 +44,7 @@ function App() {
               <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 */}
@@ -61,6 +65,7 @@ function App() {
         </BrowserRouter>
       </AuthProvider>
     </QueryClientProvider>
+    </ThemeProvider>
   );
 }
 

frontend/src/api/client.ts

@@ -126,12 +126,55 @@ export const analysisApi = {
   },
 };
 
+// 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

frontend/src/components/Layout.tsx

@@ -1,9 +1,11 @@
 import { Outlet, NavLink, useNavigate } from 'react-router-dom';
 import { useAuth } from '../context/AuthContext';
-import { Search, Layers, BarChart3, Info, Users, LogOut } from 'lucide-react';
+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 = () => {
@@ -15,6 +17,7 @@ export function Layout() {
     { 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' },
   ];
 
@@ -23,7 +26,7 @@ export function Layout() {
   }
 
   return (
-    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-indigo-950">
+    <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">
@@ -63,6 +66,13 @@ export function Layout() {
 
             {/* 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>

frontend/src/components/ProtectedRoute.tsx

@@ -12,7 +12,7 @@ export function ProtectedRoute({ children, requireAdmin = false }: ProtectedRout
 
   if (isLoading) {
     return (
-      <div className="min-h-screen bg-gradient-to-br from-bg-dark to-indigo-950 flex items-center justify-center">
+      <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>
     );

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/index.css

@@ -2,6 +2,26 @@
 @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;
@@ -15,7 +35,7 @@ body {
 }
 
 ::-webkit-scrollbar-track {
-  background: #1e293b;
+  background: var(--color-bg-card);
 }
 
 ::-webkit-scrollbar-thumb {
@@ -30,5 +50,5 @@ body {
 /* Selection */
 ::selection {
   background: rgba(99, 102, 241, 0.3);
-  color: #f8fafc;
+  color: var(--color-text-primary);
 }

frontend/src/pages/Analysis.tsx

@@ -1,7 +1,7 @@
 import { useState } from 'react';
 import { useMutation } from '@tanstack/react-query';
-import { analysisApi } from '../api/client';
-import { Search, CheckCircle, AlertCircle, Clock, FileText } from 'lucide-react';
+import { analysisApi, exportApi } from '../api/client';
+import { Search, CheckCircle, AlertCircle, Clock, FileText, Download } from 'lucide-react';
 import type { CompanyAnalysis } from '../types';
 
 export function Analysis() {
@@ -106,9 +106,27 @@ export function Analysis() {
           {/* Analysis Content */}
           {result.success && result.analysis && (
             <div className="bg-bg-card/60 backdrop-blur-lg border border-primary/15 rounded-2xl p-6">
-              <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
-                AI Analysis Results
-              </h3>
+              <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}

frontend/src/pages/Analytics.tsx

@@ -2,22 +2,50 @@ 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, XAxis, YAxis, Tooltip, ResponsiveContainer, Legend } from 'recharts';
+import { PieChart, Pie, Cell, BarChart, Bar, LineChart, Line, XAxis, YAxis, Tooltip, ResponsiveContainer, Legend } from 'recharts';
 
 const COLORS = ['#6366f1', '#0ea5e9', '#10b981', '#f59e0b', '#ef4444', '#8b5cf6', '#ec4899', '#14b8a6'];
 
 export function AnalyticsPage() {
   const [days, setDays] = useState(30);
 
-  const { data, isLoading, isError } = useQuery({
+  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="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 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>
     );
   }
@@ -33,15 +61,18 @@ export function AnalyticsPage() {
         <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">Database Not Connected</span>
+            <span className="font-semibold">Unable to Load Analytics</span>
           </div>
           <p className="text-text-secondary">
-            Set <code className="bg-bg-card px-2 py-1 rounded">USE_DATABASE=true</code> in your .env file to enable analytics tracking.
+            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>
-        </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>Analytics features require storing analysis results in PostgreSQL for historical tracking.</span>
+          <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>
     );
@@ -163,6 +194,114 @@ export function AnalyticsPage() {
           </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="#94a3b8" fontSize={12} />
+                      <YAxis stroke="#94a3b8" fontSize={12} />
+                      <Tooltip
+                        contentStyle={{
+                          backgroundColor: '#1e293b',
+                          border: '1px solid rgba(99, 102, 241, 0.3)',
+                          borderRadius: '8px',
+                        }}
+                        labelStyle={{ color: '#f8fafc' }}
+                      />
+                      <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="#94a3b8" fontSize={12} />
+                      <YAxis stroke="#94a3b8" fontSize={12} />
+                      <Tooltip
+                        contentStyle={{
+                          backgroundColor: '#1e293b',
+                          border: '1px solid rgba(99, 102, 241, 0.3)',
+                          borderRadius: '8px',
+                        }}
+                        labelStyle={{ color: '#f8fafc' }}
+                      />
+                      <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>
   );
 }

frontend/src/pages/Batch.tsx

@@ -114,9 +114,21 @@ export function Batch() {
 
       {/* 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>Batch analysis failed. Please try again.</span>
+        <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>
+          <button
+            onClick={() => mutation.reset()}
+            className="ml-7 mt-2 text-sm text-primary hover:text-primary-dark underline"
+          >
+            Dismiss
+          </button>
         </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

@@ -31,7 +31,7 @@ export function Login() {
   };
 
   return (
-    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-indigo-950 flex items-center justify-center px-4">
+    <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">

frontend/src/pages/Register.tsx

@@ -40,7 +40,7 @@ export function Register() {
   };
 
   return (
-    <div className="min-h-screen bg-gradient-to-br from-bg-dark to-indigo-950 flex items-center justify-center px-4">
+    <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">

frontend/tailwind.config.js

@@ -4,6 +4,7 @@ export default {
     "./index.html",
     "./src/**/*.{js,ts,jsx,tsx}",
   ],
+  darkMode: 'class',
   theme: {
     extend: {
       colors: {
@@ -16,15 +17,15 @@ export default {
         warning: '#f59e0b',
         error: '#ef4444',
         bg: {
-          dark: '#0f172a',
-          card: '#1e293b',
-          'card-hover': '#334155',
+          dark: 'var(--color-bg-dark)',
+          card: 'var(--color-bg-card)',
+          'card-hover': 'var(--color-bg-card-hover)',
         },
         text: {
-          primary: '#f8fafc',
-          secondary: '#94a3b8',
+          primary: 'var(--color-text-primary)',
+          secondary: 'var(--color-text-secondary)',
         },
-        border: '#334155',
+        border: 'var(--color-border)',
       },
     },
   },

requirements.txt

@@ -15,3 +15,6 @@ pandas
 bcrypt
 PyJWT
 slowapi
+apscheduler
+boto3
+reportlab