Critical (1-3):
- Replace hardcoded admin credentials with secure auto-generation (seed.py)
- Enforce SECRET_KEY configuration, fail in production if missing (config.py)
- Add Zip Slip and Zip Bomb protection to all ZIP import services
High/Medium (4-9):
- Add 50MB file size limit and extension whitelist to evidence uploads
- Configure CORS origins via environment variable instead of hardcoded
- Migrate JWT storage from localStorage to HttpOnly cookies (frontend+backend)
- Add rate limiting (5/min) on login endpoint via slowapi
- Replace generic dict payloads with Pydantic schemas (mass assignment)
Medium (10-17):
- Check is_active on login to prevent disabled users from authenticating
- Sanitize exception messages in API responses (system, data_sources)
- Escape LIKE wildcards in all ilike search filters across 8 routers
- Run Docker container as non-root user (appuser)
- Make MINIO_SECURE configurable via environment variable
- Add password complexity policy (12+ chars, upper/lower/digit/special)
- Implement JWT token revocation via in-memory blacklist + reduce TTL to 15min
- Replace xml.etree with defusedxml to prevent Billion Laughs attacks
Low (18-20):
- Add security headers to Nginx (CSP, X-Frame-Options, HSTS-ready, etc.)
- Disable Swagger UI/ReDoc/OpenAPI in production
- Restrict /health endpoint to internal networks via Nginx ACL
Also: rewrite install.sh as interactive wizard for guided deployment,
fix test-from-template validation error (technique_id UUID vs MITRE ID)
- Rewrite D3FEND import to use tactic-level APIs for reliable technique
fetching with proper ontology IRIs, descriptions, and tactic assignments
- Fix D3FEND technique URLs to use canonical IRI (no more 404s)
- All 255 D3FEND techniques now have descriptions from the official API
- Change Deactivate button color to red in template management table
- Add custom Aegis logo and favicon replacing default Vite assets
- Remove unused old API parsing code and clean up fallback list
- Make D3FEND defense cards clickable with expandable details and external link
- Fix D3FEND URLs to use PascalCase technique names matching the ontology
- Remove duplicate Import Atomic Red Team from System page (use Data Sources)
- Add bulk Activate All / Deactivate All buttons with confirmation modal
- Fix template admin list to show both active and inactive templates
- Add PATCH /test-templates/bulk-activate backend endpoint
- Auto-seed data sources on container startup via entrypoint.sh
- Fix SigmaHQ, CALDERA, GTFOBins import issues
- Register D3FEND sync handler in data sources router
- Add CIS Controls v8 compliance framework import
- Expand Test Catalog source filters (CALDERA, LOLBAS, GTFOBins)
- Campaign Generate from Threat Actor now opens actor selector modal
- Add coverage snapshot creation button to Comparison page
- Update README with accurate data source and feature documentation
T-106: Create test_workflow_service.py with state-machine transitions for the complete test lifecycle (draft -> red_executing -> blue_evaluating -> in_review -> validated/rejected), dual validation by Red/Blue leads, and reopen capability with field cleanup.
T-107: Update status_service.py to use detection_result from Blue Team instead of legacy result field, and differentiate between partial progress (some validated) vs all-in-progress states.
T-108: Create atomic_import_service.py that downloads the Atomic Red Team repo as a ZIP (avoiding API rate limits), parses all atomics YAML files, and creates idempotent TestTemplate records mapped to MITRE techniques.
Includes validation tests for all three tasks (19 checks total).
- Add intel_service.py: RSS feed scanner for threat intelligence
Searches CISA, NIST NVD, SANS ISC, BleepingComputer, The Hacker News,
Krebs on Security for mentions of MITRE technique IDs and names
- New intel items stored in intel_items table with URL deduplication
- Techniques with new intel flagged with review_required=True
- Add POST /system/run-intel-scan endpoint (admin only)
- Register weekly intel scan job in APScheduler (every 7 days)
- Audit log records each scan execution with summary stats
- Update README with new endpoint and project structure
- Add MITRE sync service via TAXII 2.0 with GitHub fallback
- Upsert attack-pattern objects into techniques table (691 techniques)
- Detect name/description changes and flag review_required on re-sync
- Add APScheduler background job running every 24h
- Add POST /system/sync-mitre endpoint (admin only)
- Add GET /system/scheduler-status endpoint (admin only)
- Configure logging for scheduler and sync visibility
- Update README with new endpoints and project structure
- Add Pydantic schemas for Technique, Test and Evidence
- Add CRUD endpoints for Techniques (list with filters, detail, create, update, review)
- Add CRUD endpoints for Tests (create, detail, update, validate, reject)
- Add evidence upload with SHA-256 integrity and presigned download URLs
- Add MinIO/S3 storage client with bucket auto-creation on startup
- Add status_service to recalculate technique coverage from test results
- Add require_any_role RBAC dependency for multi-role authorization
- Update README with API endpoints reference and project structure
Implements all database models for the Aegis platform with full
Alembic migration support.
Models created:
- User: Authentication with role-based access control
- Technique: MITRE ATT&CK techniques with coverage status tracking
- Test: Security tests with validation workflow (draft/review/validated)
- Evidence: File metadata for test evidence (stored in MinIO)
- IntelItem: Threat intelligence items linked to techniques
- AuditLog: System-wide audit trail with JSONB details
Enumerations:
- TechniqueStatus: not_evaluated, in_progress, validated, partial, etc.
- TestState: draft, in_review, validated, rejected
- TestResult: detected, not_detected, partially_detected
Services:
- audit_service.py: log_action() helper for audit logging
All models include proper foreign key relationships and PostgreSQL
enum types are managed correctly in migrations (create/drop).
