feat(audit): V2 mail render + 5 new findings (B4/B5/B6/B7/B8) + LLM-Plausibility-Phase

Mail Render V2 (compliance/services/mail_render_v2/) — 11-Modul-Subpackage
das einen einheitlichen Audit-Mail-Output erzeugt mit:
  - Header + KPI-Kacheln (Score / Findings / Docs / Vendors)
  - TOC + Sprung-Links
  - 3-Bucket-Trennung: Kritische Befunde / Manuelle Prüfung / Interne Reminder
  - Cookie-Inventar (Name·Vendor·Kategorie·Speicherdauer·Löschfrist·Sitzland·Quelle·Status)
  - Sofortmaßnahmen-Aggregator ("Sitzland ergänzen für 11 Cookies")
  - 24 Legacy-Wrappers — alle alten build_*_html in V2-Sections
  - Scope-Filter: FIN/GOV/MED/INS/EDU/LEG aus Berichten wenn nicht relevant
  - Hint/Action-Dedup: keine doppelten Sätze pro Card mehr
Aktiviert via env MAIL_RENDER_V2=true (Default: legacy renderer).

5 neue deterministische Findings als Phase D-2b/B4/B5/B6/B7/B8:

  B4 vendor_consistency_check — Cross-Doc-Provider-Widerspruch
     (Elli: DSE nennt Vertex AI für Chatbot, /de/cookies nennt Iadvize → HIGH).
     6 Service-Types: chatbot/analytics/tag_manager/pixel/cdn/cmp.

  B5 ai_act_transparency_check — AI Act Art. 50 Transparenzpflicht
     (Elli: Vertex AI vorhanden ohne Pre-Chat-Disclosure → HIGH).
     Plus B5-Erweiterung: Rechtsgrundlage Art-6-Abs-1-lit-f bei AI → MED
     (Einwilligung empfehlen).

  B6 cross_doc_dpo_check — DPO in DSE genannt, nicht im Impressum (LOW).

  B7 doc_staleness_check — Datum-Extraktion aus DSE/AGB/Nutzungsbedingungen.
     Cap: AGB/NB 3y, DSE 2y. Älter → MEDIUM (Elli NB Stand 2018 → HIGH).

  B8 cmp_fingerprint_check — Banner detected, aber CMP-Provider generic
     (kein Usercentrics/OneTrust/Cookiebot/etc → MED).

  B3-Erweiterung detect_intra_doc_contradictions — Widersprüchliche
     Speicherdauer im SELBEN Doc (Elli: Logfile 7d vs 30d → HIGH).

LLM-Plausibility-Phase (Phase D-2b, finding_plausibility_check.py):
  - Läuft AFTER MC pipeline, BEFORE D3 render
  - Prompt mit Beispiel-IDs + 3-Phase-Mapping: exact-ID / position-fallback /
    fuzzy-tail-match
  - Stempelt llm_title / llm_severity / llm_recommendation / llm_drop auf
    jeden FAIL CheckItem
  - V2-Render zeigt "🤖 LLM-Plausibility:" Box pro Finding wenn gestempelt
  - KNOWN ISSUE: qwen3:30b-a3b liefert oft empty content auf format='json' +
    8000-char-excerpt prompts. Pipeline läuft mit stamped=0 weiter. Task #16.

Coverage gegen Elli Ground Truth (zeroclaw/docs/ground-truth/elli_eco_2026-06-06.json,
13 expected findings via WebFetch-Agent-Crawl):
  - 4/4 HIGH-Findings ✓ (COOKIE-CONSENT-UX-001 + WIDERRUFSBELEHRUNG-001 +
    VENDOR-CONSISTENCY-001 + AI-ACT-TRANSPARENCY-001)
  - 4/6 MEDIUM ✓
  - 2/3 LOW ✓
  - Total: 10/13 = 77% (Sprung von 4/13 = 31%)

Restliche 3 Gaps als Task #17: IMPRESSUM-001 (multi-entity USt-IdNr),
TRANSFER-001 (Vendor-Mechanismus DPF/SCC), TH-RETENTION-002 (AI-Retention
pro Datenkategorie).

V2-Mail-Preview in Mailpit: 'v2all@local.test' Subject '[V2 ALL] ELLI'.
Backend healthy, B1+B3+B4+B5+B6+B7+B8 alle live im Orchestrator.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-06-06 21:19:49 +02:00
parent c2c8783fee
commit d0e3621192
27 changed files with 4426 additions and 3 deletions
@@ -0,0 +1,78 @@
"""B4 wiring — Cross-Doc Vendor-Consistency check + HTML block.
Activated after B1+B3 in the orchestrator. The check itself is
deterministic (no LLM); it scans DSE + cookie texts for known
service providers per service type and flags every mismatch.
The mail renderer reads `state["vendor_consistency_findings"]` and
`state["vendor_consistency_html"]` directly — no further wiring.
"""
from __future__ import annotations
import html
import logging
from compliance.services.vendor_consistency_check import (
check_vendor_consistency,
)
logger = logging.getLogger(__name__)
def run_b4(state: dict) -> None:
findings = check_vendor_consistency(state)
state["vendor_consistency_findings"] = findings
if not findings:
return
state["vendor_consistency_html"] = _render(findings)
logger.info(
"B4 Vendor-Consistency: %d findings (HIGH=%d, MEDIUM=%d)",
len(findings),
sum(1 for f in findings if (f.get("severity") or "") == "HIGH"),
sum(1 for f in findings if (f.get("severity") or "") == "MEDIUM"),
)
def _render(findings: list[dict]) -> str:
rows = []
for f in findings:
sev = (f.get("severity") or "").upper()
color = "#dc2626" if sev == "HIGH" else "#f59e0b"
dse = ", ".join(f.get("dse_providers") or []) or "<em></em>"
cookie = ", ".join(f.get("cookie_providers") or []) or "<em></em>"
rows.append(
"<tr>"
f"<td style='padding:6px 10px;border-bottom:1px solid #e5e7eb;'>"
f"{html.escape((f.get('service_type') or '').replace('_',' ').title())}"
"</td>"
f"<td style='padding:6px 10px;border-bottom:1px solid #e5e7eb;'>"
f"{dse}</td>"
f"<td style='padding:6px 10px;border-bottom:1px solid #e5e7eb;'>"
f"{cookie}</td>"
f"<td style='padding:6px 10px;border-bottom:1px solid #e5e7eb;"
f"color:{color};font-weight:600;'>"
f"{sev} {html.escape(f.get('severity_reason') or '')}</td>"
"</tr>"
)
return (
"<div style='margin:24px 0;padding:16px;border-left:4px solid #dc2626;"
"background:#fff1f2;border-radius:4px;'>"
"<h2 style='margin:0 0 8px;color:#991b1b;font-size:16px;'>"
"VENDOR-CONSISTENCY-001 — Vendor-Konsistenz DSE ↔ Cookies</h2>"
"<p style='margin:0 0 8px;font-size:14px;color:#3f3f46;'>"
f"<strong>{len(findings)}</strong> Provider-Widersprüche zwischen "
"Datenschutzerklärung und Cookie-Seite. Beispiel Elli: "
"DSE = Vertex AI für Chatbot, Cookies-Seite = Iadvize.</p>"
"<table style='width:100%;border-collapse:collapse;font-size:13px;"
"margin-top:8px;background:#fff;'>"
"<thead><tr style='background:#f1f5f9;'>"
"<th style='text-align:left;padding:6px 10px;'>Service-Typ</th>"
"<th style='text-align:left;padding:6px 10px;'>In DSE</th>"
"<th style='text-align:left;padding:6px 10px;'>Auf Cookies-Seite</th>"
"<th style='text-align:left;padding:6px 10px;'>Severity</th>"
"</tr></thead>"
f"<tbody>{''.join(rows)}</tbody>"
"</table>"
"</div>"
)