Benjamin Admin
c52dbdb8f1
feat(rag): optimize RAG pipeline — JSON-Mode, CoT, Hybrid Search, Re-Ranking, Cross-Reg Dedup, chunk 1024
...
CI/CD / go-lint (push) Has been skipped
CI/CD / python-lint (push) Has been skipped
CI/CD / nodejs-lint (push) Has been skipped
CI/CD / test-go-ai-compliance (push) Failing after 42s
CI/CD / test-python-backend-compliance (push) Successful in 1m38s
CI/CD / test-python-document-crawler (push) Successful in 20s
CI/CD / test-python-dsms-gateway (push) Successful in 17s
CI/CD / validate-canonical-controls (push) Successful in 10s
CI/CD / Deploy (push) Has been skipped
Phase 1 (LLM Quality):
- Add format=json to all Ollama payloads (obligation_extractor, control_generator, citation_backfill)
- Add Chain-of-Thought analysis steps to Pass 0a/0b system prompts
Phase 2 (Retrieval Quality):
- Hybrid search via Qdrant Query API with RRF fusion + automatic text index (legal_rag.go)
- Fallback to dense-only search if Query API unavailable
- Cross-encoder re-ranking with BGE Reranker v2 (RERANK_ENABLED=false by default)
- CPU-only PyTorch dependency to keep Docker image small
Phase 3 (Data Layer):
- Cross-regulation dedup pass (threshold 0.95) links controls across regulations
- DedupResult.link_type field distinguishes dedup_merge vs cross_regulation
- Chunk size defaults updated 512/50 → 1024/128 for new ingestions only
- Existing collections and controls are NOT affected
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-03-21 11:49:43 +01:00
Benjamin Admin
95fcba34cd
fix(quality): Ruff/CVE/TS-Fixes, 104 neue Tests, Complexity-Refactoring
...
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-ai-compliance (push) Failing after 30s
CI / test-python-backend-compliance (push) Successful in 30s
CI / test-python-document-crawler (push) Successful in 21s
CI / test-python-dsms-gateway (push) Successful in 17s
- Ruff: 144 auto-fixes (unused imports, == None → is None), F821/F811/F841 manuell
- CVEs: python-multipart>=0.0.22, weasyprint>=68.0, pillow>=12.1.1, npm audit fix (0 vulns)
- TS: 5 tote Drafting-Engine-Dateien entfernt, allowed-facts/sanitizer/StepHeader/context fixes
- Tests: +104 (ISMS 58, Evidence 18, VVT 14, Generation 14) → 1449 passed
- Refactoring: collect_ci_evidence (F→A), row_to_response (E→A), extract_requirements (E→A)
- Dead Code: pca-platform, 7 Go-Handler, dsr_api.py, duplicate Schemas entfernt
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-03-07 19:00:33 +01:00
Benjamin Admin
34fc8dc654
feat: 6 Dokumentations-Module auf 100% — VVT Backend, Filter, PDF-Export
...
Phase 1 — VVT Backend (localStorage → API):
- migrations/006_vvt.sql: Neue Tabellen (vvt_organization, vvt_activities, vvt_audit_log)
- compliance/db/vvt_models.py: SQLAlchemy-Models für alle VVT-Tabellen
- compliance/api/vvt_routes.py: Vollständiger CRUD-Router (10 Endpoints)
- compliance/api/__init__.py: VVT-Router registriert
- compliance/api/schemas.py: VVT Pydantic-Schemas ergänzt
- app/(sdk)/sdk/vvt/page.tsx: API-Client + camelCase↔snake_case Mapping,
localStorage durch persistente DB-Calls ersetzt (POST/PUT/DELETE/GET)
- tests/test_vvt_routes.py: 18 Tests (alle grün)
Phase 3 — Document Generator PDF-Export:
- document-generator/page.tsx: "Als PDF exportieren"-Button funktioniert jetzt
via window.print() + Print-Window mit korrektem HTML
- Fallback-Banner wenn Template-Service (breakpilot-core) nicht erreichbar
Phase 4 — Source Policy erweiterte Filter:
- SourcesTab.tsx: source_type-Filter (Rechtlich / Leitlinien / Vorlagen / etc.)
- PIIRulesTab.tsx: category-Filter (E-Mail / Telefon / IBAN / etc.)
- source_policy_router.py: Backend-Endpoints unterstützen jetzt source_type
und category als Query-Parameter
- requirements.txt: reportlab==4.2.5 ergänzt (fehlende Audit-PDF-Dependency)
Phase 2 — Training (Migration-Skripte):
- scripts/apply_training_migrations.sh: SSH-Skript für Mac Mini
- scripts/apply_vvt_migration.sh: Vollständiges Deploy-Skript für VVT
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-03-02 17:14:58 +01:00
Benjamin Admin
e6d666b89b
feat: Vorbereitung-Module auf 100% — Persistenz, Backend-Services, UCCA Frontend
...
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-ai-compliance (push) Successful in 37s
CI / test-python-backend-compliance (push) Successful in 32s
CI / test-python-document-crawler (push) Successful in 22s
CI / test-python-dsms-gateway (push) Successful in 18s
Phase A: PostgreSQL State Store (sdk_states Tabelle, InMemory-Fallback)
Phase B: Modules dynamisch vom Backend, Scope DB-Persistenz, Source Policy State
Phase C: UCCA Frontend (3 Seiten, Wizard, RiskScoreGauge), Obligations Live-Daten
Phase D: Document Import (PDF/LLM/Gap-Analyse), System Screening (SBOM/OSV.dev)
Phase E: Company Profile CRUD mit Audit-Logging
Phase F: Tests (Python + TypeScript), flow-data.ts DB-Tabellen aktualisiert
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-03-02 11:04:31 +01:00
Benjamin Boenisch
4435e7ea0a
Initial commit: breakpilot-compliance - Compliance SDK Platform
...
Services: Admin-Compliance, Backend-Compliance,
AI-Compliance-SDK, Consent-SDK, Developer-Portal,
PCA-Platform, DSMS
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-02-11 23:47:28 +01:00
