feat: DSK/BfDI RAG-Ingest, TOM-Control-Library 180, Risk-Engine-Spec, RAG-Query-Optimierung
All checks were successful
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 37s
CI / test-python-document-crawler (push) Successful in 23s
CI / test-python-dsms-gateway (push) Successful in 21s

- Crawler erweitert: +26 neue Dokumente (DSK KP 1-20, SDM V3.1, BfDI Loeschkonzept, BayLDA TOM-Checkliste)
- RAG-Queries optimiert: 18 Queries mit EDPB/DSK/WP-Referenzen fuer besseres Retrieval
- Chat-Route: queryRAG nutzt jetzt Collection + Query-Boost aus DOCUMENT_RAG_CONFIG
- TOM Control Library: 180 Controls in 12 Domaenen (ISO Annex-A Style, tom_controls_v1.json)
- Risk Engine Spec: Impact/Likelihood 0-10, Score 0-100, 4 Tiers, Loeschfristen-Engine
- Soul-Files: DSK-Kurzpapiere, SDM V3.1, BfDI als primaere deutsche Quellen
- Manifest CSV: eu_de_privacy_manifest.csv mit Lizenz-Ampel (gruen/gelb/rot)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-03-05 12:03:57 +01:00
parent 3913931d5b
commit 3707ffe799
7 changed files with 8938 additions and 24 deletions

View File

@@ -8,7 +8,9 @@
import { NextRequest, NextResponse } from 'next/server'
import { queryRAG } from '@/lib/sdk/drafting-engine/rag-query'
import { DOCUMENT_RAG_CONFIG } from '@/lib/sdk/drafting-engine/rag-config'
import { readSoulFile } from '@/lib/sdk/agents/soul-reader'
import type { ScopeDocumentType } from '@/lib/sdk/compliance-scope-types'
const OLLAMA_URL = process.env.OLLAMA_URL || 'http://host.docker.internal:11434'
const LLM_MODEL = process.env.COMPLIANCE_LLM_MODEL || 'qwen2.5vl:32b'
@@ -40,8 +42,10 @@ export async function POST(request: NextRequest) {
return NextResponse.json({ error: 'Message is required' }, { status: 400 })
}
// 1. Query RAG for legal context
const ragContext = await queryRAG(message)
// 1. Query RAG for legal context (use type-specific collection + query boost if available)
const ragConfig = documentType ? DOCUMENT_RAG_CONFIG[documentType as ScopeDocumentType] : undefined
const ragQuery = ragConfig ? `${ragConfig.query} ${message}` : message
const ragContext = await queryRAG(ragQuery, 3, ragConfig?.collection)
// 2. Build system prompt with mode-specific instructions + state projection
const soulPrompt = await readSoulFile('drafting-agent')