deploy: security hardening, multi-model support, S3 storage, analytics, CI improvements (70 commits) #4

Merged
0xWheatyz merged 70 commits from leeworks-agents/SPARC:main into main 2026-03-31 11:53:47 +00:00
3 changed files with 162 additions and 0 deletions
Showing only changes of commit a4db9439f5 - Show all commits
+6
View File
@@ -40,3 +40,9 @@ JWT_SECRET=your-secure-jwt-secret-change-in-production
# When USE_CACHE=true: check database for cached responses before making API calls # 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) # When USE_CACHE=false: always make fresh API calls (still stores results in database)
USE_CACHE=true USE_CACHE=true
# ---- 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
+17
View File
@@ -670,8 +670,25 @@ def _run_batch_job(job_id: str, companies: list[str], max_workers: int):
progress=100, progress=100,
result_json=_json.dumps(batch_response.model_dump(), default=str), 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: except Exception as e:
db.update_job(job_id, status="failed", error=str(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"]) @app.post("/analyze/batch/async", response_model=JobStatus, tags=["Analysis"])
+139
View File
@@ -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,
})