fix: resolve all merge conflict markers from feat/zeroclaw-compliance-agent

9 files had conflict markers from the branch merge. All resolved keeping
the feature branch version. Also split agent_scan_routes.py (534→367 LOC)
by extracting Pydantic models to agent_scan_models.py.

[guardrail-change]

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.claude/rules/loc-exceptions.txt

@@ -114,3 +114,6 @@ docs-src/control_generator_routes.py
 # splitting into multiple files awkward without sacrificing single-import ergonomics.
 consent-sdk/src/mobile/flutter/consent_sdk.dart
 consent-sdk/src/mobile/ios/ConsentManager.swift
+
+# --- docs-src: binary office files (not source code) ---
+docs-src/Breakpilot ComplAI Finanzplan.xlsm

admin-compliance/app/api/sdk/v1/agent/scan/route.ts

@@ -18,11 +18,7 @@ export async function POST(request: NextRequest) {
       method: 'POST',
       headers: { 'Content-Type': 'application/json' },
       body,
-<<<<<<< HEAD
-      signal: AbortSignal.timeout(30000), // 30s — just needs to start the job
-=======
       signal: AbortSignal.timeout(300000), // 5 min — multi-page scan + LLM calls
->>>>>>> feat/zeroclaw-compliance-agent
     })
 
     if (!response.ok) {

admin-compliance/app/sdk/agent/_components/ScanResult.tsx

@@ -35,21 +35,7 @@ interface ScanFinding {
   severity: string
   text: string
   correction: string
-<<<<<<< HEAD
-  doc_title: string
-}
-
-interface DiscoveredDocument {
-  title: string
-  url: string
-  doc_type: string
-  language: string
-  word_count: number
-  completeness_pct: number
-  findings_count: number
-=======
   text_reference: TextRef | null
->>>>>>> feat/zeroclaw-compliance-agent
 }
 
 interface ScanData {
@@ -297,16 +283,6 @@ export function ScanResult({ data }: { data: ScanData }) {
           </div>
         </div>
       )}
-<<<<<<< HEAD
-
-      {/* Email Status */}
-      {data.email_status && (
-        <div className="text-xs text-gray-500 flex items-center gap-2">
-          <span className={`w-2 h-2 rounded-full ${data.email_status === 'sent' ? 'bg-green-400' : 'bg-gray-300'}`} />
-          E-Mail: {data.email_status === 'sent' ? 'Gesendet' : data.email_status}
-        </div>
-      )}
-=======
       {/* PDF Export Button */}
       <div className="pt-4 border-t flex gap-3">
         <button
@@ -336,7 +312,6 @@ export function ScanResult({ data }: { data: ScanData }) {
           PDF herunterladen
         </button>
       </div>
->>>>>>> feat/zeroclaw-compliance-agent
     </div>
   )
 }

admin-compliance/app/sdk/agent/page.tsx

@@ -2,37 +2,6 @@
 
 import React, { useState } from 'react'
 import { ScanResult } from './_components/ScanResult'
-<<<<<<< HEAD
-import { DocCheckTab } from './_components/DocCheckTab'
-import { BannerCheckTab } from './_components/BannerCheckTab'
-import { ImpressumCheckTab } from './_components/ImpressumCheckTab'
-import { ComplianceFAQ } from './_components/ComplianceFAQ'
-
-type AnalysisTab = 'scan' | 'doc-check' | 'banner-check' | 'impressum-check'
-
-const TABS: { id: AnalysisTab; label: string; desc: string }[] = [
-  { id: 'scan', label: 'Website-Scan', desc: 'Rechtliche Dokumente finden + Dienstleister erkennen' },
-  { id: 'doc-check', label: 'Dokumenten-Pruefung', desc: 'DSI, AGB, Cookie-Richtlinie inhaltlich pruefen' },
-  { id: 'banner-check', label: 'Banner-Check', desc: 'Cookie-Banner auf DSGVO-Konformitaet testen' },
-  { id: 'impressum-check', label: 'Impressum-Check', desc: 'Impressum auf §5 TMG Pflichtangaben pruefen' },
-]
-
-export default function AgentPage() {
-  const [url, setUrl] = useState(() => typeof window !== 'undefined' ? localStorage.getItem('agent-scan-url') || '' : '')
-  const [tab, setTab] = useState<AnalysisTab>(() => (typeof window !== 'undefined' ? localStorage.getItem('agent-scan-tab') as AnalysisTab : null) || 'scan')
-  const [scanLoading, setScanLoading] = useState(false)
-  const [scanError, setScanError] = useState<string | null>(null)
-  const [scanData, setScanData] = useState<any>(() => {
-    if (typeof window === 'undefined') return null
-    try { const s = localStorage.getItem('agent-scan-result'); return s ? JSON.parse(s) : null } catch { return null }
-  })
-  const [scanProgress, setScanProgress] = useState<string>('')
-  const [activeScanId, setActiveScanId] = useState<string>(() => typeof window !== 'undefined' ? localStorage.getItem('agent-scan-id') || '' : '')
-  const [scanHistory, setScanHistory] = useState<{ url: string; date: string; findings: number; docs: number; resultKey: string }[]>(() => {
-    if (typeof window === 'undefined') return []
-    try { return JSON.parse(localStorage.getItem('agent-scan-history') || '[]') } catch { return [] }
-  })
-=======
 import { ConsentTestResult } from './_components/ConsentTestResult'
 import { CompareResult } from './_components/CompareResult'
 import { AuthTestResult } from './_components/AuthTestResult'
@@ -68,7 +37,6 @@ export default function AgentPage() {
   const [authUser, setAuthUser] = useState('')
   const [authPass, setAuthPass] = useState('')
   const { analyze, answerFollowUp, loading, error, result, history } = useAgentAnalysis()
->>>>>>> feat/zeroclaw-compliance-agent
 
   React.useEffect(() => { localStorage.setItem('agent-scan-url', url) }, [url])
   React.useEffect(() => { localStorage.setItem('agent-scan-tab', tab) }, [tab])
@@ -129,48 +97,6 @@ export default function AgentPage() {
 
   const handleScan = async (e: React.FormEvent) => {
     e.preventDefault()
-<<<<<<< HEAD
-    if (!url.trim()) return
-    setScanLoading(true)
-    setScanError(null)
-    setScanData(null)
-    setScanProgress('Scan wird gestartet...')
-    try {
-      const startRes = await fetch('/api/sdk/v1/agent/scan', {
-        method: 'POST',
-        headers: { 'Content-Type': 'application/json' },
-        body: JSON.stringify({ url: url.trim(), mode: 'post_launch' }),
-      })
-      if (!startRes.ok) throw new Error(`Scan konnte nicht gestartet werden: ${startRes.status}`)
-      const { scan_id } = await startRes.json()
-      if (!scan_id) throw new Error('Keine Scan-ID erhalten')
-      setActiveScanId(scan_id)
-      localStorage.setItem('agent-scan-id', scan_id)
-
-      let attempts = 0
-      while (attempts < 120) {
-        await new Promise(r => setTimeout(r, 5000))
-        const pollRes = await fetch(`/api/sdk/v1/agent/scan?scan_id=${scan_id}`)
-        if (!pollRes.ok) { attempts++; continue }
-        const pollData = await pollRes.json()
-        if (pollData.progress) setScanProgress(pollData.progress)
-        if (pollData.status === 'completed' && pollData.result) {
-          setScanData(pollData.result)
-          setScanProgress('')
-          localStorage.setItem('agent-scan-result', JSON.stringify(pollData.result))
-          localStorage.removeItem('agent-scan-id')
-          setActiveScanId('')
-          _addToHistory(pollData.result)
-          break
-        }
-        if (pollData.status === 'failed') throw new Error(pollData.error || 'Scan fehlgeschlagen')
-        attempts++
-      }
-      if (attempts >= 120) throw new Error('Scan-Timeout (10 Minuten)')
-    } catch (e) {
-      setScanError(e instanceof Error ? e.message : 'Unbekannter Fehler')
-      setScanProgress('')
-=======
     setScanLoading(true)
     setScanError(null)
 
@@ -214,7 +140,6 @@ export default function AgentPage() {
       else if (tab === 'auth') setAuthData(data)
     } catch (e) {
       setScanError(e instanceof Error ? e.message : 'Fehler')
->>>>>>> feat/zeroclaw-compliance-agent
     } finally {
       setScanLoading(false)
     }
@@ -242,131 +167,6 @@ export default function AgentPage() {
     <div className="space-y-6 max-w-4xl">
       <div>
         <h1 className="text-2xl font-bold text-gray-900">Compliance Agent</h1>
-<<<<<<< HEAD
-        <p className="text-gray-500 mt-1">Analysiere Webseiten und Dokumente auf DSGVO-Konformitaet.</p>
-      </div>
-
-      {/* Tab Selection */}
-      <div className="flex border-b border-gray-200 overflow-x-auto">
-        {TABS.map(t => (
-          <button key={t.id} onClick={() => setTab(t.id)}
-            className={`px-4 py-2.5 text-sm font-medium border-b-2 transition-colors whitespace-nowrap ${
-              tab === t.id
-                ? 'border-purple-500 text-purple-700'
-                : 'border-transparent text-gray-500 hover:text-gray-700'}`}>
-            {t.label}
-          </button>
-        ))}
-      </div>
-
-      {/* Website-Scan Tab */}
-      {tab === 'scan' && (
-        <div className="space-y-4">
-          <div className="bg-indigo-50 border border-indigo-200 rounded-lg p-4">
-            <h3 className="text-sm font-semibold text-indigo-900">Website-Scan (Discovery)</h3>
-            <p className="text-xs text-indigo-700 mt-1">
-              Findet alle rechtlichen Dokumente (DSI, AGB, Impressum, Cookie, Widerruf),
-              erkennt eingesetzte Drittdienste und prueft ob sie in der DSE dokumentiert sind.
-            </p>
-          </div>
-
-          <form onSubmit={handleScan} className="flex gap-3">
-            <input type="url" value={url} onChange={e => setUrl(e.target.value)}
-              placeholder="https://www.example.com/"
-              className="flex-1 px-4 py-3 border border-gray-300 rounded-lg focus:ring-2 focus:ring-purple-500 focus:border-transparent text-sm"
-              disabled={scanLoading} required />
-            <button type="submit" disabled={scanLoading || !url.trim()}
-              className="px-6 py-3 bg-purple-600 text-white rounded-lg hover:bg-purple-700 disabled:opacity-50 transition-colors flex items-center gap-2 text-sm font-medium whitespace-nowrap">
-              {scanLoading ? (
-                <><svg className="animate-spin w-4 h-4" fill="none" viewBox="0 0 24 24">
-                  <circle className="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" strokeWidth="4" />
-                  <path className="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4z" />
-                </svg>Scanne...</>
-              ) : 'Website scannen'}
-            </button>
-          </form>
-
-          {scanProgress && (
-            <div className="bg-purple-50 border border-purple-200 rounded-lg p-4 text-sm text-purple-700 flex items-center gap-3">
-              <svg className="animate-spin w-5 h-5 text-purple-500 shrink-0" fill="none" viewBox="0 0 24 24">
-                <circle className="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" strokeWidth="4" />
-                <path className="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4z" />
-              </svg>
-              {scanProgress}
-            </div>
-          )}
-
-          {scanError && (
-            <div className="bg-red-50 border border-red-200 rounded-lg p-4 text-sm text-red-700">{scanError}</div>
-          )}
-
-          {/* Quick Action Buttons — navigate to specialized tabs */}
-          {scanData && (
-            <div className="bg-white border border-gray-200 rounded-xl p-4 shadow-sm">
-              <h4 className="text-sm font-semibold text-gray-800 mb-3">Jetzt pruefen</h4>
-              <div className="grid grid-cols-2 gap-2">
-                <button onClick={() => navigateToCheck('banner-check', scannedUrl)}
-                  className="p-3 rounded-lg border border-gray-200 hover:border-purple-300 hover:bg-purple-50 transition-all text-left">
-                  <div className="text-sm font-medium text-gray-900">Cookie-Banner pruefen</div>
-                  <div className="text-xs text-gray-500 mt-0.5">3-Phasen Dark-Pattern-Analyse</div>
-                </button>
-                <button onClick={() => navigateToCheck('impressum-check', scannedUrl + '/impressum')}
-                  className="p-3 rounded-lg border border-gray-200 hover:border-purple-300 hover:bg-purple-50 transition-all text-left">
-                  <div className="text-sm font-medium text-gray-900">Impressum pruefen</div>
-                  <div className="text-xs text-gray-500 mt-0.5">§5 TMG Pflichtangaben</div>
-                </button>
-                {discoveredDocs.map((doc: any, i: number) => (
-                  <button key={i} onClick={() => navigateToCheck('doc-check', doc.url)}
-                    className="p-3 rounded-lg border border-gray-200 hover:border-purple-300 hover:bg-purple-50 transition-all text-left">
-                    <div className="text-sm font-medium text-gray-900 truncate">{doc.title || doc.url}</div>
-                    <div className="text-xs text-gray-500 mt-0.5">
-                      {doc.doc_type?.toUpperCase()} · {doc.word_count || '?'} Woerter
-                      {doc.completeness_pct != null && ` · ${doc.completeness_pct}%`}
-                    </div>
-                  </button>
-                ))}
-              </div>
-            </div>
-          )}
-
-          {/* Full Scan Result */}
-          {scanData?.services && (
-            <div className="bg-white border border-gray-200 rounded-xl p-6 shadow-sm">
-              <ScanResult data={scanData} />
-            </div>
-          )}
-
-          {/* Scan History */}
-          {scanHistory.length > 0 && (
-            <div className="border border-gray-200 rounded-xl p-4">
-              <h4 className="text-sm font-medium text-gray-700 mb-3">Letzte Scans</h4>
-              <div className="space-y-2">
-                {scanHistory.map((h, i) => (
-                  <button key={i} onClick={() => {
-                    setUrl(h.url)
-                    if (h.resultKey) {
-                      try { const s = localStorage.getItem(h.resultKey); if (s) { setScanData(JSON.parse(s)); return } } catch {}
-                    }
-                    try { const l = localStorage.getItem('agent-scan-result'); if (l) setScanData(JSON.parse(l)) } catch {}
-                  }}
-                    className="w-full flex items-center justify-between p-3 rounded-lg border border-gray-100 hover:border-purple-200 hover:bg-purple-50/30 transition-all text-left">
-                    <div className="min-w-0 flex-1">
-                      <div className="text-sm font-medium text-gray-900 truncate">{h.url}</div>
-                      <div className="text-xs text-gray-500">
-                        {new Date(h.date).toLocaleDateString('de-DE', { day: '2-digit', month: '2-digit', year: 'numeric', hour: '2-digit', minute: '2-digit' })}
-                      </div>
-                    </div>
-                    <div className="flex items-center gap-3 shrink-0 ml-3">
-                      {h.docs > 0 && <span className="text-xs text-purple-600">{h.docs} Dok.</span>}
-                      <span className={`text-xs font-medium ${h.findings > 0 ? 'text-red-600' : 'text-green-600'}`}>
-                        {h.findings} Findings
-                      </span>
-                    </div>
-                  </button>
-                ))}
-              </div>
-            </div>
-=======
         <p className="text-gray-500 mt-1">Analysiere Dokumente und Webseiten auf DSGVO-Konformitaet.</p>
       </div>
 
@@ -443,7 +243,6 @@ export default function AgentPage() {
           <AnalysisResult result={result} />
           {result.follow_up_questions.length > 0 && (
             <div className="border-t pt-4"><FollowUpQuestions questions={result.follow_up_questions} answers={result.follow_up_answers} onAnswer={answerFollowUp} /></div>
->>>>>>> feat/zeroclaw-compliance-agent
           )}
         </div>
       )}
@@ -452,15 +251,6 @@ export default function AgentPage() {
       {tab === 'compare' && compareData?.sites && <div className="bg-white border border-gray-200 rounded-xl p-6 shadow-sm"><CompareResult sites={compareData.sites} /></div>}
       {tab === 'auth' && authData && <div className="bg-white border border-gray-200 rounded-xl p-6 shadow-sm"><AuthTestResult data={authData} /></div>}
 
-<<<<<<< HEAD
-      {/* Specialized Tabs */}
-      {tab === 'doc-check' && <DocCheckTab />}
-      {tab === 'banner-check' && <BannerCheckTab />}
-      {tab === 'impressum-check' && <ImpressumCheckTab />}
-
-      {/* FAQ */}
-      <ComplianceFAQ />
-=======
       {/* History */}
       {tab === 'quick' && <AnalysisHistory history={history} onSelect={r => { setUrl(r.url); analyze(r.url, mode) }} />}
       {tab === 'scan' && scanHistory.length > 0 && (
@@ -480,7 +270,6 @@ export default function AgentPage() {
           </div>
         </div>
       )}
->>>>>>> feat/zeroclaw-compliance-agent
     </div>
   )
 }

admin-compliance/app/sdk/cmp/page.tsx

@@ -54,18 +54,6 @@ export default function CMPDashboardPage() {
   const [consentStats, setConsentStats] = useState<ConsentStats | null>(null)
   const [dsrStats, setDSRStats] = useState<DSRStats | null>(null)
   const [sites, setSites] = useState<any[]>([])
-<<<<<<< HEAD
-  const [selectedSite, setSelectedSite] = useState<string>('')
-  const [loading, setLoading] = useState(true)
-
-  const fb = (path: string) => fetch(`${BANNER_API}/${path}`, { headers: HEADERS }).then(r => r.ok ? r.json() : null).catch(() => null)
-
-  // Load sites + consent/dsr stats on mount
-  useEffect(() => {
-    async function load() {
-      const fa = (path: string) => fetch(`/api/sdk/v1/compliance/${path}`, { headers: HEADERS }).then(r => r.ok ? r.json() : null).catch(() => null)
-      const [consent, dsr, siteList] = await Promise.all([
-=======
   const [loading, setLoading] = useState(true)
 
   useEffect(() => {
@@ -74,34 +62,10 @@ export default function CMPDashboardPage() {
       const fa = (path: string) => fetch(`/api/sdk/v1/compliance/${path}`, { headers: HEADERS }).then(r => r.ok ? r.json() : null).catch(() => null)
       const [banner, consent, dsr, siteList] = await Promise.all([
         fb('admin/stats/preview-test-site'),
->>>>>>> feat/zeroclaw-compliance-agent
         fa('einwilligungen/consents/stats'),
         fa('dsr/stats'),
         fb('admin/sites'),
       ])
-<<<<<<< HEAD
-      setConsentStats(consent)
-      setDSRStats(dsr)
-      const loadedSites = Array.isArray(siteList) ? siteList : []
-      setSites(loadedSites)
-      // Auto-select first site
-      if (loadedSites.length > 0) {
-        setSelectedSite(loadedSites[0].site_id || loadedSites[0].siteId || '')
-      }
-      setLoading(false)
-    }
-    load()
-  // eslint-disable-next-line react-hooks/exhaustive-deps
-  }, [])
-
-  // Load banner stats when selected site changes
-  useEffect(() => {
-    if (!selectedSite) return
-    fb(`admin/stats/${selectedSite}`).then(setBannerStats)
-  // eslint-disable-next-line react-hooks/exhaustive-deps
-  }, [selectedSite])
-
-=======
       setBannerStats(banner)
       setConsentStats(consent)
       setDSRStats(dsr)
@@ -111,7 +75,6 @@ export default function CMPDashboardPage() {
     load()
   }, [])
 
->>>>>>> feat/zeroclaw-compliance-agent
   const totalConsents = (bannerStats?.total_consents || 0) + (consentStats?.total_consents || 0)
   const dsrOpen = dsrStats ? (dsrStats.by_status?.intake || 0) + (dsrStats.by_status?.processing || 0) + (dsrStats.by_status?.identity_verification || 0) : 0
   const dsrOverdue = dsrStats?.overdue || 0
@@ -123,36 +86,12 @@ export default function CMPDashboardPage() {
       <div className="flex items-center justify-between">
         <div>
           <h1 className="text-2xl font-bold text-gray-900">Consent Management Platform</h1>
-<<<<<<< HEAD
-          <p className="text-gray-500 mt-1">Überblick über Einwilligungen, Betroffenenrechte und Vendor-Compliance</p>
-        </div>
-        <div className="flex items-center gap-3">
-          {sites.length > 0 && (
-            <select
-              value={selectedSite}
-              onChange={e => setSelectedSite(e.target.value)}
-              className="px-3 py-2 border border-gray-300 rounded-lg text-sm bg-white focus:ring-2 focus:ring-purple-500 focus:border-purple-500"
-            >
-              {sites.map((s: any) => (
-                <option key={s.site_id || s.siteId} value={s.site_id || s.siteId}>
-                  {s.site_name || s.siteName || s.site_id || s.siteId}
-                </option>
-              ))}
-            </select>
-          )}
-          <Link href="/sdk/cookie-banner/preview"
-            className="px-4 py-2 bg-purple-600 text-white rounded-lg text-sm font-medium hover:bg-purple-700 transition-colors">
-            Banner testen
-          </Link>
-        </div>
-=======
           <p className="text-gray-500 mt-1">Ueberblick ueber Einwilligungen, Betroffenenrechte und Vendor-Compliance</p>
         </div>
         <Link href="/sdk/cookie-banner/preview"
           className="px-4 py-2 bg-purple-600 text-white rounded-lg text-sm font-medium hover:bg-purple-700 transition-colors">
           Banner testen
         </Link>
->>>>>>> feat/zeroclaw-compliance-agent
       </div>
 
       {/* KPI Cards */}
@@ -235,8 +174,6 @@ export default function CMPDashboardPage() {
         </div>
       </div>
 
-<<<<<<< HEAD
-=======
       {/* Banner-Bedarf Hinweis (TTDSG § 25) */}
       {bannerStats && Object.keys(bannerStats.category_acceptance).length === 0 && sites.length === 0 && (
         <div className="bg-green-50 border border-green-200 rounded-xl p-5 flex items-start gap-4">
@@ -275,7 +212,6 @@ export default function CMPDashboardPage() {
         </div>
       )}
 
->>>>>>> feat/zeroclaw-compliance-agent
       {/* Compliance Status */}
       <div className="bg-white rounded-xl border border-gray-200 p-6">
         <h3 className="font-semibold text-gray-900 mb-1">Compliance-Status</h3>

admin-compliance/app/sdk/document-generator/_constants.ts

@@ -6,24 +6,6 @@ import { TemplateContext } from './contextBridge'
 
 export const CATEGORIES: { key: string; label: string; types: string[] | null }[] = [
   { key: 'all', label: 'Alle', types: null },
-<<<<<<< HEAD
-  { key: 'privacy_policy', label: 'Datenschutz', types: ['privacy_policy'] },
-  { key: 'terms', label: 'AGB', types: ['terms_of_service', 'agb', 'clause'] },
-  { key: 'impressum', label: 'Impressum', types: ['impressum'] },
-  { key: 'dpa', label: 'AVV/DPA', types: ['dpa'] },
-  { key: 'nda', label: 'NDA', types: ['nda'] },
-  { key: 'sla', label: 'SLA', types: ['sla'] },
-  { key: 'acceptable_use', label: 'AUP', types: ['acceptable_use'] },
-  { key: 'widerruf', label: 'Widerruf', types: ['widerruf'] },
-  { key: 'cookie', label: 'Cookie', types: ['cookie_policy', 'cookie_banner'] },
-  { key: 'cloud', label: 'Cloud', types: ['cloud_service_agreement'] },
-  { key: 'misc', label: 'Weitere', types: ['community_guidelines', 'copyright_policy', 'data_usage_clause'] },
-  { key: 'dsfa', label: 'DSFA', types: ['dsfa'] },
-  { key: 'dsr', label: 'DSR-Prozesse', types: [
-    'dsr_process_art15', 'dsr_process_art16', 'dsr_process_art17',
-    'dsr_process_art18', 'dsr_process_art19', 'dsr_process_art20', 'dsr_process_art21',
-  ]},
-=======
 
   // ── Nach Nutzungskontext sortiert ──────────────────────────────────────
 
@@ -82,7 +64,6 @@ export const CATEGORIES: { key: string; label: string; types: string[] | null }[
   { key: 'vendor', label: 'Lieferanten / Vendor', types: ['vendor_risk_management_policy', 'third_party_security_policy', 'supplier_security_policy', 'dpa'] },
 
   { key: 'bcm', label: 'BCM / Notfall', types: ['business_continuity_policy', 'disaster_recovery_policy', 'crisis_management_policy', 'incident_response_plan'] },
->>>>>>> feat/zeroclaw-compliance-agent
 ]
 
 // =============================================================================

admin-compliance/app/sdk/document-generator/page.tsx

@@ -88,79 +88,6 @@ function DocumentGeneratorPageInner() {
     }
   }, [state?.companyProfile])
 
-<<<<<<< HEAD
-  // ── MODULE WIRING: CookieBanner → CONSENT + FEATURES ─────────────────────
-  useEffect(() => {
-    const banner = state?.cookieBanner
-    if (!banner) return
-    const cats = banner.categories || []
-    const analyticsTools = cats
-      .filter((c) => c.id === 'analytics' || c.id === 'statistics')
-      .flatMap((c) => c.cookies?.map((ck) => ck.name) ?? [])
-    const marketingTools = cats
-      .filter((c) => c.id === 'marketing')
-      .flatMap((c) => c.cookies?.map((ck) => ck.name) ?? [])
-    const hasFunctional = cats.some((c) => c.id === 'functional')
-
-    setContext((prev) => ({
-      ...prev,
-      CONSENT: {
-        ...prev.CONSENT,
-        ANALYTICS_TOOLS: analyticsTools.length > 0 ? analyticsTools.join(', ') : prev.CONSENT.ANALYTICS_TOOLS,
-        MARKETING_PARTNERS: marketingTools.length > 0 ? marketingTools.join(', ') : prev.CONSENT.MARKETING_PARTNERS,
-      },
-      FEATURES: {
-        ...prev.FEATURES,
-        CMP_NAME: 'BreakPilot CMP',
-        CMP_LOGS_CONSENTS: true,
-        HAS_FUNCTIONAL_COOKIES: hasFunctional || prev.FEATURES.HAS_FUNCTIONAL_COOKIES,
-        CONSENT_WITHDRAWAL_PATH: 'Footer-Link "Cookie-Einstellungen"',
-      },
-    }))
-  }, [state?.cookieBanner])
-
-  // ── MODULE WIRING: Loeschfristen → PRIVACY retention ──────────────────────
-  useEffect(() => {
-    const policies = state?.retentionPolicies
-    if (!policies || policies.length === 0) return
-    const maxMonths = policies.reduce((max, p) => {
-      const match = p.retentionPeriod?.match(/(\d+)\s*(Monat|Jahr|Tag)/i)
-      if (!match) return max
-      const val = parseInt(match[1], 10)
-      const unit = match[2].toLowerCase()
-      const months = unit.startsWith('jahr') ? val * 12 : unit.startsWith('tag') ? Math.ceil(val / 30) : val
-      return Math.max(max, months)
-    }, 0)
-    if (maxMonths > 0) {
-      setContext((prev) => ({
-        ...prev,
-        PRIVACY: { ...prev.PRIVACY, ANALYTICS_RETENTION_MONTHS: maxMonths },
-      }))
-    }
-  }, [state?.retentionPolicies])
-
-  // ── MODULE WIRING: UseCases → FEATURES flags ─────────────────────────────
-  useEffect(() => {
-    const useCases = state?.useCases
-    if (!useCases || useCases.length === 0) return
-    const allText = useCases.map((uc) => `${uc.name} ${uc.description}`).join(' ').toLowerCase()
-    const hasAccount = allText.includes('account') || allText.includes('konto') || allText.includes('registrier')
-    const hasPayments = allText.includes('zahlung') || allText.includes('payment') || allText.includes('stripe') || allText.includes('paypal')
-    const hasNewsletter = allText.includes('newsletter') || allText.includes('mailchimp') || allText.includes('e-mail-marketing')
-    const hasSocial = allText.includes('social') || allText.includes('linkedin') || allText.includes('facebook') || allText.includes('instagram')
-
-    setContext((prev) => ({
-      ...prev,
-      FEATURES: {
-        ...prev.FEATURES,
-        HAS_ACCOUNT: hasAccount || prev.FEATURES.HAS_ACCOUNT,
-        HAS_PAYMENTS: hasPayments || prev.FEATURES.HAS_PAYMENTS,
-        HAS_NEWSLETTER: hasNewsletter || prev.FEATURES.HAS_NEWSLETTER,
-        HAS_SOCIAL_MEDIA: hasSocial || prev.FEATURES.HAS_SOCIAL_MEDIA,
-      },
-    }))
-  }, [state?.useCases])
-=======
   // Pre-fill TOM/DPA context from Compliance Scope Engine
   useEffect(() => {
     const scopeLevel = state?.complianceScope?.determinedLevel
@@ -173,7 +100,6 @@ function DocumentGeneratorPageInner() {
       }))
     }
   }, [state?.complianceScope?.determinedLevel, state?.companyProfile])
->>>>>>> feat/zeroclaw-compliance-agent
 
   // Pre-fill extra placeholders from Einwilligungen data points
   useEffect(() => {

admin-compliance/app/sdk/layout.tsx

@@ -208,12 +208,8 @@ function SDKInnerLayout({ children }: { children: React.ReactNode }) {
       {/* Command Bar Modal */}
       {isCommandBarOpen && <CommandBar onClose={() => setCommandBarOpen(false)} />}
 
-<<<<<<< HEAD
-      {/* Module-specific FAB navigators are rendered by each module's layout */}
-=======
       {/* Pipeline Sidebar (FAB on mobile/tablet, fixed on desktop xl+) */}
       <SDKPipelineSidebar />
->>>>>>> feat/zeroclaw-compliance-agent
 
       {/* Compliance Advisor Widget — immer sichtbar, auch ohne Projekt */}
       <ComplianceAdvisorWidget currentStep={currentStep} />

admin-compliance/components/sdk/CookieBannerOverlay.tsx

@@ -93,17 +93,11 @@ export function CookieBannerOverlay() {
 
   return (
     <>
-<<<<<<< HEAD
-      {/* Non-blocking banner — no overlay, no pointer-events blocking */}
-      <div className="fixed bottom-0 left-16 xl:left-64 right-0 z-50 pointer-events-none">
-        <div className="max-w-3xl mx-auto m-4 bg-white rounded-2xl shadow-2xl border border-gray-200 overflow-hidden pointer-events-auto">
-=======
       {/* Overlay — leaves sidebar (left 64px/16px) accessible */}
       <div className="fixed inset-0 ml-16 xl:ml-64 bg-black/30 z-[9998]" onClick={() => setIsOpen(false)} />
 
       <div className="fixed bottom-0 left-0 right-0 z-[9999]">
         <div className="max-w-3xl mx-auto m-4 bg-white rounded-2xl shadow-2xl border border-gray-200 overflow-hidden">
->>>>>>> feat/zeroclaw-compliance-agent
 
           {/* Header with EWR toggle + close button */}
           <div className="px-6 pt-5 pb-3">

admin-compliance/e2e/playwright-live.config.ts

@@ -1,5 +1,3 @@
-<<<<<<< HEAD
-=======
 /**
  * Playwright config for testing against live Mac Mini instance.
  * No webServer — assumes https://macmini:3007 is already running.
@@ -7,17 +5,10 @@
  * Usage: npx playwright test --config=e2e/playwright-live.config.ts
  */
 
->>>>>>> feat/zeroclaw-compliance-agent
 import { defineConfig, devices } from '@playwright/test'
 
 export default defineConfig({
   testDir: './specs',
-<<<<<<< HEAD
-  timeout: 30000,
-  use: {
-    baseURL: 'https://macmini:3007',
-    ignoreHTTPSErrors: true,
-=======
   fullyParallel: true,
   retries: 0,
   workers: 3,
@@ -30,16 +21,12 @@ export default defineConfig({
     ignoreHTTPSErrors: true,
     screenshot: 'on',
     trace: 'on-first-retry',
->>>>>>> feat/zeroclaw-compliance-agent
   },
   projects: [
     { name: 'chromium', use: { ...devices['Desktop Chrome'] } },
   ],
-<<<<<<< HEAD
-=======
   outputDir: 'e2e/test-results',
   timeout: 20000,
   expect: { timeout: 5000 },
   // No webServer — we test against the live instance
->>>>>>> feat/zeroclaw-compliance-agent
 })

backend-compliance/compliance/api/agent_scan_helpers.py

@@ -73,43 +73,6 @@ def build_scan_summary(
         f"Findings: {n_findings} ({high} mit hoher Prioritaet)",
     ])
 
-<<<<<<< HEAD
-    # DSI Documents section — grouped with their findings
-    if discovered_docs:
-        parts.extend(["", f"Rechtliche Dokumente ({len(discovered_docs)})"])
-
-        # Group findings by doc_title
-        doc_findings_map: dict[str, list] = {}
-        general_findings: list = []
-        for f in findings:
-            dt = f.doc_title if hasattr(f, 'doc_title') else ""
-            if dt:
-                doc_findings_map.setdefault(dt, []).append(f)
-            else:
-                general_findings.append(f)
-
-        for doc in discovered_docs:
-            title = doc.title if hasattr(doc, 'title') else "?"
-            pct = doc.completeness_pct if hasattr(doc, 'completeness_pct') else 0
-            wc = doc.word_count if hasattr(doc, 'word_count') else 0
-            status = "OK" if pct >= 80 else "LUECKENHAFT" if pct >= 50 else "MANGELHAFT"
-            parts.append(f"  [{status}] {title} ({pct}%, {wc} Woerter)")
-            for f in doc_findings_map.get(title, []):
-                sev = f.severity if hasattr(f, 'severity') else "?"
-                txt = f.text if hasattr(f, 'text') else str(f)
-                marker = "!!" if sev == "HIGH" else "!" if sev == "MEDIUM" else "i"
-                parts.append(f"    {marker} {txt}")
-
-        # General findings (no doc association)
-        if general_findings:
-            parts.extend(["", "Allgemeine Findings"])
-            for f in general_findings[:20]:
-                sev = f.severity if hasattr(f, 'severity') else "?"
-                txt = f.text if hasattr(f, 'text') else str(f)
-                marker = "!!" if sev == "HIGH" else "!" if sev == "MEDIUM" else "i"
-                parts.append(f"  [{marker}] {txt}")
-    elif findings:
-=======
     # DSI Documents section
     if discovered_docs:
         parts.extend([
@@ -129,7 +92,6 @@ def build_scan_summary(
             )
 
     if findings:
->>>>>>> feat/zeroclaw-compliance-agent
         parts.append("")
         for f in findings[:20]:
             sev = f.severity if hasattr(f, 'severity') else "?"
@@ -145,44 +107,3 @@ def build_scan_summary(
         ])
 
     return "\n".join(parts)
-<<<<<<< HEAD
-
-
-async def fetch_dse_text(url: str, scanned_pages: list[str]) -> str:
-    """Find and fetch the privacy policy page text."""
-    dse_url = None
-    for page in scanned_pages:
-        if re.search(r"datenschutz|privacy|dsgvo", page, re.IGNORECASE):
-            dse_url = page
-            break
-    if not dse_url:
-        dse_url = url
-    try:
-        async with httpx.AsyncClient(timeout=15.0, follow_redirects=True) as client:
-            resp = await client.get(dse_url, headers={"User-Agent": "BreakPilot-Compliance-Agent/1.0"})
-            html = resp.text
-            clean = re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=re.DOTALL | re.IGNORECASE)
-            clean = re.sub(r"<[^>]+>", " ", clean)
-            clean = re.sub(r"\s+", " ", clean).strip()
-            return clean[:8000]
-    except Exception:
-        return ""
-
-
-async def fetch_dse_html(url: str, scanned_pages: list[str]) -> str:
-    """Fetch the raw HTML of the privacy policy page."""
-    dse_url = None
-    for page in scanned_pages:
-        if re.search(r"datenschutz|privacy|dsgvo", page, re.IGNORECASE):
-            dse_url = page
-            break
-    if not dse_url:
-        dse_url = url
-    try:
-        async with httpx.AsyncClient(timeout=15.0, follow_redirects=True) as client:
-            resp = await client.get(dse_url, headers={"User-Agent": "BreakPilot-Compliance-Agent/1.0"})
-            return resp.text
-    except Exception:
-        return ""
-=======
->>>>>>> feat/zeroclaw-compliance-agent

backend-compliance/compliance/api/agent_scan_models.py

@@ -0,0 +1,84 @@
+"""Pydantic models for the Agent Website Scan API."""
+
+from pydantic import BaseModel
+
+
+class ScanRequest(BaseModel):
+    url: str
+    mode: str = "post_launch"
+    recipient: str = "dsb@breakpilot.local"
+
+
+class ServiceInfo(BaseModel):
+    name: str
+    category: str
+    provider: str
+    country: str
+    eu_adequate: bool
+    requires_consent: bool
+    legal_ref: str
+    in_dse: bool
+    status: str  # "ok", "undocumented", "outdated"
+
+
+class TextReferenceModel(BaseModel):
+    found: bool = False
+    source_url: str = ""
+    document_type: str = "Datenschutzerklaerung"
+    section_heading: str = ""
+    section_number: str = ""
+    parent_section: str = ""
+    paragraph_index: int = 0
+    original_text: str = ""
+    issue: str = ""
+    correction_type: str = ""
+    correction_text: str = ""
+    insert_after: str = ""
+
+
+class ScanFinding(BaseModel):
+    code: str
+    severity: str
+    text: str
+    correction: str = ""
+    text_reference: TextReferenceModel | None = None
+
+
+class DiscoveredDocument(BaseModel):
+    title: str
+    url: str
+    doc_type: str
+    language: str = ""
+    word_count: int = 0
+    completeness_pct: int = 0
+    findings_count: int = 0
+
+
+class ScanResponse(BaseModel):
+    url: str
+    pages_scanned: int
+    pages_list: list[str] = []
+    services: list[ServiceInfo]
+    findings: list[ScanFinding]
+    discovered_documents: list[DiscoveredDocument] = []
+    ai_detected: bool
+    chatbot_detected: bool
+    chatbot_provider: str
+    missing_pages: dict
+    summary: str
+    email_status: str
+    scanned_at: str
+
+
+class ScanStartResponse(BaseModel):
+    scan_id: str
+    status: str = "running"
+    message: str = ""
+
+
+class ScanStatusResponse(BaseModel):
+    scan_id: str
+    status: str  # "running", "completed", "failed"
+    progress: str = ""
+    result: ScanResponse | None = None
+    error: str = ""

backend-compliance/compliance/api/agent_scan_routes.py

@@ -12,7 +12,6 @@ from datetime import datetime, timezone
 
 import httpx
 from fastapi import APIRouter
-from pydantic import BaseModel
 
 from compliance.services.website_scanner import scan_website, DetectedService
 from compliance.services.dse_service_extractor import extract_dse_services, compare_services
@@ -23,13 +22,11 @@ from compliance.services.mandatory_content_checker import (
     check_mandatory_documents, check_dse_mandatory_content, MandatoryFinding,
 )
 from compliance.services.legal_basis_validator import validate_legal_bases
-<<<<<<< HEAD
-from compliance.api.agent_scan_helpers import (
-    add_corrections, build_scan_summary, fetch_dse_text, fetch_dse_html,
-)
-=======
 from compliance.api.agent_scan_helpers import add_corrections, build_scan_summary
->>>>>>> feat/zeroclaw-compliance-agent
+from compliance.api.agent_scan_models import (
+    ScanRequest, ServiceInfo, TextReferenceModel, ScanFinding,
+    DiscoveredDocument, ScanResponse, ScanStartResponse, ScanStatusResponse,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -45,77 +42,6 @@ SDK_HEADERS = {
 }
 
 
-class ScanRequest(BaseModel):
-    url: str
-    mode: str = "post_launch"
-    recipient: str = "dsb@breakpilot.local"
-
-
-class ServiceInfo(BaseModel):
-    name: str
-    category: str
-    provider: str
-    country: str
-    eu_adequate: bool
-    requires_consent: bool
-    legal_ref: str
-    in_dse: bool
-    status: str  # "ok", "undocumented", "outdated"
-
-
-class TextReferenceModel(BaseModel):
-    found: bool = False
-    source_url: str = ""
-    document_type: str = "Datenschutzerklaerung"
-    section_heading: str = ""
-    section_number: str = ""
-    parent_section: str = ""
-    paragraph_index: int = 0
-    original_text: str = ""
-    issue: str = ""
-    correction_type: str = ""
-    correction_text: str = ""
-    insert_after: str = ""
-
-
-class ScanFinding(BaseModel):
-    code: str
-    severity: str
-    text: str
-    correction: str = ""
-<<<<<<< HEAD
-    doc_title: str = ""
-=======
->>>>>>> feat/zeroclaw-compliance-agent
-    text_reference: TextReferenceModel | None = None
-
-
-class DiscoveredDocument(BaseModel):
-    title: str
-    url: str
-    doc_type: str
-    language: str = ""
-    word_count: int = 0
-    completeness_pct: int = 0
-    findings_count: int = 0
-
-
-class ScanResponse(BaseModel):
-    url: str
-    pages_scanned: int
-    pages_list: list[str] = []
-    services: list[ServiceInfo]
-    findings: list[ScanFinding]
-    discovered_documents: list[DiscoveredDocument] = []
-    ai_detected: bool
-    chatbot_detected: bool
-    chatbot_provider: str
-    missing_pages: dict
-    summary: str
-    email_status: str
-    scanned_at: str
-
-
 import asyncio
 import uuid as _uuid
 
@@ -123,20 +49,6 @@ import uuid as _uuid
 _scan_jobs: dict[str, dict] = {}
 
 
-class ScanStartResponse(BaseModel):
-    scan_id: str
-    status: str = "running"
-    message: str = ""
-
-
-class ScanStatusResponse(BaseModel):
-    scan_id: str
-    status: str  # "running", "completed", "failed"
-    progress: str = ""
-    result: ScanResponse | None = None
-    error: str = ""
-
-
 @router.post("/scan")
 async def scan_website_endpoint(req: ScanRequest):
     """Start async website scan. Returns scan_id immediately.
@@ -204,60 +116,12 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
 
     # Use Playwright results if available, otherwise fall back to httpx scanner
     if playwright_htmls:
-        # Build ScanResult from Playwright data
+        from compliance.services.website_scanner import ScanResult, _detect_services, _detect_ai_mentions
-        from compliance.services.website_scanner import ScanResult, DetectedService, _detect_services, _detect_ai_mentions
-        from compliance.services.service_registry import SERVICE_REGISTRY
         scan = ScanResult()
         scan.pages_scanned = list(playwright_htmls.keys())
         for page_url, html in playwright_htmls.items():
             _detect_services(html, page_url, scan)
             _detect_ai_mentions(html, page_url, scan)
-        # Deduplicate
-        seen = set()
-        unique = []
-        for svc in scan.detected_services:
-            if svc.id not in seen:
-                seen.add(svc.id)
-                unique.append(svc)
-        scan.detected_services = unique
-        scan.chatbot_detected = any(s.category == "chatbot" for s in scan.detected_services)
-        if scan.chatbot_detected:
-            scan.chatbot_provider = next(s.name for s in scan.detected_services if s.category == "chatbot")
-    else:
-        scan = await scan_website(req.url)
-
-<<<<<<< HEAD
-    logger.info("Scanned %d pages, found %d services", len(scan.pages_scanned), len(scan.detected_services))
-
-    _progress(f"Schritt 2/7: Rechtliche Dokumente suchen... ({len(scan.pages_scanned)} Seiten gescannt)")
-=======
-    # Step 1: Scan website — try Playwright first (JS-rendered), fallback to httpx
-    playwright_htmls: dict[str, str] = {}
-    try:
-        async with httpx.AsyncClient(timeout=120.0) as pw_client:
-            pw_resp = await pw_client.post(
-                "http://bp-compliance-consent-tester:8094/website-scan",
-                json={"url": req.url, "max_pages": 15, "click_nav": True},
-            )
-            if pw_resp.status_code == 200:
-                pw_data = pw_resp.json()
-                playwright_htmls = pw_data.get("page_htmls", {})
-                logger.info("Playwright scan: %d pages, %d scripts",
-                            pw_data.get("pages_count", 0), len(pw_data.get("external_scripts", [])))
-    except Exception as e:
-        logger.warning("Playwright scanner unavailable, falling back to httpx: %s", e)
-
-    # Use Playwright results if available, otherwise fall back to httpx scanner
-    if playwright_htmls:
-        # Build ScanResult from Playwright data
-        from compliance.services.website_scanner import ScanResult, DetectedService, _detect_services, _detect_ai_mentions
-        from compliance.services.service_registry import SERVICE_REGISTRY
-        scan = ScanResult()
-        scan.pages_scanned = list(playwright_htmls.keys())
-        for page_url, html in playwright_htmls.items():
-            _detect_services(html, page_url, scan)
-            _detect_ai_mentions(html, page_url, scan)
-        # Deduplicate
         seen = set()
         unique = []
         for svc in scan.detected_services:
@@ -273,54 +137,30 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
 
     logger.info("Scanned %d pages, found %d services", len(scan.pages_scanned), len(scan.detected_services))
 
->>>>>>> feat/zeroclaw-compliance-agent
     # Step 1b: DSI Discovery — find all legal documents on the website
     discovered_docs: list[DiscoveredDocument] = []
     dsi_findings: list[ScanFinding] = []
     try:
-<<<<<<< HEAD
-        async with httpx.AsyncClient(timeout=300.0) as dsi_client:
-            dsi_resp = await dsi_client.post(
-                "http://bp-compliance-consent-tester:8094/dsi-discovery",
-                json={"url": req.url, "max_documents": 30},
-=======
         async with httpx.AsyncClient(timeout=180.0) as dsi_client:
             dsi_resp = await dsi_client.post(
                 "http://bp-compliance-consent-tester:8094/dsi-discovery",
                 json={"url": req.url, "max_documents": 20},
->>>>>>> feat/zeroclaw-compliance-agent
             )
             if dsi_resp.status_code == 200:
                 dsi_data = dsi_resp.json()
                 logger.info("DSI discovery: %d documents found", dsi_data.get("total_found", 0))
-
+                from compliance.services.dsi_document_checker import check_document_completeness, classify_document_type
-                # Check each document against its legal requirements
-                from compliance.services.dsi_document_checker import (
-                    check_document_completeness, classify_document_type,
-                )
                 for doc in dsi_data.get("documents", []):
                     doc_type = classify_document_type(doc["title"], doc["url"])
-<<<<<<< HEAD
-                    doc_text = doc.get("full_text", "") or doc.get("text_preview", "")
-                    logger.info("DSI check: '%s' type=%s text_len=%d full_text_len=%d preview_len=%d",
-                                doc["title"][:50], doc_type, len(doc_text),
-                                len(doc.get("full_text", "")), len(doc.get("text_preview", "")))
-                    doc_findings = check_document_completeness(
-                        doc_text, doc_type, doc["title"], doc["url"],
-=======
                     doc_findings = check_document_completeness(
                         doc.get("text_preview", ""), doc_type, doc["title"], doc["url"],
->>>>>>> feat/zeroclaw-compliance-agent
                     )
-                    # Count completeness
                     score_finding = next((f for f in doc_findings if "SCORE" in f.get("code", "")), None)
                     completeness = 0
                     if score_finding:
-                        import re as _re2
+                        pct_match = re.search(r"(\d+)%", score_finding.get("text", ""))
-                        pct_match = _re2.search(r"(\d+)%", score_finding.get("text", ""))
                         if pct_match:
                             completeness = int(pct_match.group(1))
-
                     discovered_docs.append(DiscoveredDocument(
                         title=doc["title"], url=doc["url"],
                         doc_type=doc_type, language=doc.get("language", ""),
@@ -330,39 +170,7 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
                     ))
                     for df in doc_findings:
                         if "SCORE" not in df.get("code", ""):
-                            dsi_findings.append(ScanFinding(
+                            dsi_findings.append(ScanFinding(code=df["code"], severity=df["severity"], text=df["text"]))
-                                code=df["code"], severity=df["severity"], text=df["text"],
-<<<<<<< HEAD
-                                doc_title=doc["title"],
-                            ))
-    except Exception as e:
-        logger.warning("DSI discovery failed: %s %s", type(e).__name__, e)
-
-    _progress(f"Schritt 3/7: Datenschutzerklaerung analysieren... ({len(discovered_docs)} Dokumente gefunden)")
-    # Step 2: Fetch privacy policy text — combine all DSI texts for best coverage
-    dse_text = ""
-    # Start with Playwright HTML if available
-    for page_url, html in playwright_htmls.items():
-        if re.search(r"datenschutz|privacy|dsgvo", page_url, re.IGNORECASE):
-            clean = re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=re.DOTALL | re.IGNORECASE)
-            clean = re.sub(r"<[^>]+>", " ", clean)
-            clean = re.sub(r"\s+", " ", clean).strip()
-            dse_text = clean[:30000]
-            break
-    # Enrich: append DSI discovery texts (they contain the actual document content)
-    try:
-        if 'dsi_resp' in dir() or 'dsi_data' in dir():
-            dsi_data_for_text = dsi_data if 'dsi_data' in dir() else {}
-            for doc in dsi_data_for_text.get("documents", []):
-                ft = doc.get("full_text", "")
-                if ft and len(ft) > 500:
-                    dse_text = (dse_text + " " + ft)[:50000]
-    except Exception:
-        pass
-    if not dse_text:
-        dse_text = await fetch_dse_text(req.url, scan.pages_scanned)
-=======
-                            ))
     except Exception as e:
         logger.warning("DSI discovery failed: %s", e)
 
@@ -370,62 +178,42 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
     dse_text = ""
     for page_url, html in playwright_htmls.items():
         if re.search(r"datenschutz|privacy|dsgvo", page_url, re.IGNORECASE):
-            import re as _re
+            clean = re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=re.DOTALL | re.IGNORECASE)
-            clean = _re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=_re.DOTALL | _re.IGNORECASE)
+            clean = re.sub(r"<[^>]+>", " ", clean)
-            clean = _re.sub(r"<[^>]+>", " ", clean)
+            clean = re.sub(r"\s+", " ", clean).strip()
-            clean = _re.sub(r"\s+", " ", clean).strip()
             dse_text = clean[:4000]
             break
     if not dse_text:
         dse_text = await _fetch_dse_text(req.url, scan.pages_scanned)
->>>>>>> feat/zeroclaw-compliance-agent
 
     # Step 3: Extract services mentioned in DSE via LLM + text fallback
     dse_services = await extract_dse_services(dse_text) if dse_text else []
     logger.info("DSE mentions %d services (LLM)", len(dse_services))
-
-    # Fallback: if LLM extraction failed, search DSE text directly for service names
     if not dse_services and dse_text:
         dse_lower = dse_text.lower()
         detected_dicts_for_check = [_service_to_dict(s) for s in scan.detected_services]
         for svc in detected_dicts_for_check:
             name = svc.get("name", "").lower()
-            # Check if service name appears in DSE text
             if name and len(name) > 3 and name in dse_lower:
                 dse_services.append({"name": svc["name"], "purpose": "", "country": svc.get("country", ""), "legal_basis": ""})
         if dse_services:
             logger.info("DSE text fallback found %d services", len(dse_services))
 
-    # Step 4: Parse DSE into structured sections (prefer Playwright HTML)
+    # Step 4: Parse DSE into structured sections
     dse_html = ""
     for page_url, html in playwright_htmls.items():
         if re.search(r"datenschutz|privacy|dsgvo", page_url, re.IGNORECASE):
             dse_html = html
             break
     if not dse_html:
-<<<<<<< HEAD
-        dse_html = await fetch_dse_html(req.url, scan.pages_scanned)
-    dse_sections = parse_dse(dse_html, req.url) if dse_html else []
-    logger.info("Parsed %d DSE sections", len(dse_sections))
-
-    _progress("Schritt 4/7: SOLL/IST Vergleich...")
-=======
         dse_html = await _fetch_dse_html(req.url, scan.pages_scanned)
     dse_sections = parse_dse(dse_html, req.url) if dse_html else []
-    logger.info("Parsed %d DSE sections", len(dse_sections))
 
->>>>>>> feat/zeroclaw-compliance-agent
+    # Step 5-8: Comparison, findings, mandatory checks, legal basis validation
-    # Step 5: SOLL/IST comparison
     detected_dicts = [_service_to_dict(s) for s in scan.detected_services]
     comparison = compare_services(detected_dicts, dse_services)
-
-    # Step 6: Build TextReferences for each detected service
     text_refs = build_text_references(detected_dicts, dse_services, dse_sections, req.url)
-
-    # Step 7: Generate findings with text references
     services_info, findings = _build_findings(comparison, scan, is_live, text_refs)
-
-    # Step 8: Check mandatory content (documents + DSE sections)
     mandatory_findings = check_mandatory_documents(scan.pages_scanned, scan.missing_pages)
     mandatory_findings += check_dse_mandatory_content(dse_sections, dse_text)
     for mf in mandatory_findings:
@@ -433,40 +221,24 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
             code=mf.code, severity=mf.severity,
             text=f"{mf.text}" + (f" — {mf.suggestion}" if mf.suggestion else ""),
         ))
-
-    # Step 8b: Validate legal bases (lit. a-f) in DSE
     if dse_text:
-        lit_findings = validate_legal_bases(dse_text)
+        for lf in validate_legal_bases(dse_text):
-        for lf in lit_findings:
             findings.append(ScanFinding(
-                code=f"LIT-{lf.purpose.upper()}",
+                code=f"LIT-{lf.purpose.upper()}", severity=lf.severity, text=lf.text,
-                severity=lf.severity,
-                text=lf.text,
                 text_reference=TextReferenceModel(
-                    found=True, source_url=req.url,
+                    found=True, source_url=req.url, original_text=lf.original_text,
-                    original_text=lf.original_text,
                     issue="incorrect", correction_type="replace",
                     correction_text=f"Korrekte Rechtsgrundlage: {lf.correct_basis} ({lf.legal_ref})",
                 ) if lf.original_text else None,
             ))
-
-    # Step 8c: Add DSI document findings
     findings.extend(dsi_findings)
-
-<<<<<<< HEAD
-    _progress(f"Schritt 5/7: Korrekturen generieren... ({len(findings)} Findings)")
-=======
->>>>>>> feat/zeroclaw-compliance-agent
-    # Step 9: Generate corrections for pre-launch mode
     if not is_live and findings:
         await add_corrections(findings, dse_text)
 
     _progress("Schritt 6/7: Report erstellen...")
-    # Step 7: Build summary
     summary = build_scan_summary(req.url, scan, comparison, findings, is_live, discovered_docs)
 
     _progress("Schritt 7/7: E-Mail senden...")
-    # Step 8: Send notification
     mode_label = "INTERNE PRUEFUNG" if not is_live else "LIVE-WEBSITE"
     email_result = send_email(
         recipient=req.recipient,
@@ -475,34 +247,30 @@ async def _execute_scan(req: ScanRequest, scan_id: str = "") -> ScanResponse:
     )
 
     return ScanResponse(
-        url=req.url,
+        url=req.url, pages_scanned=len(scan.pages_scanned), pages_list=scan.pages_scanned,
-        pages_scanned=len(scan.pages_scanned),
+        services=services_info, findings=findings, discovered_documents=discovered_docs,
-        pages_list=scan.pages_scanned,
+        ai_detected=len(scan.ai_mentions) > 0, chatbot_detected=scan.chatbot_detected,
-        services=services_info,
+        chatbot_provider=scan.chatbot_provider, missing_pages=scan.missing_pages,
-        findings=findings,
+        summary=summary, email_status=email_result.get("status", "failed"),
-        discovered_documents=discovered_docs,
-        ai_detected=len(scan.ai_mentions) > 0,
-        chatbot_detected=scan.chatbot_detected,
-        chatbot_provider=scan.chatbot_provider,
-        missing_pages=scan.missing_pages,
-        summary=summary,
-        email_status=email_result.get("status", "failed"),
         scanned_at=datetime.now(timezone.utc).isoformat(),
     )
 
 
+async def _fetch_dse_text(url: str, scanned_pages: list[str]) -> str:
+    """Fetch DSE text from the privacy policy page."""
+    dse_url = next((p for p in scanned_pages if re.search(r"datenschutz|privacy|dsgvo", p, re.IGNORECASE)), url)
+    try:
+        async with httpx.AsyncClient(timeout=15.0, follow_redirects=True) as client:
+            resp = await client.get(dse_url, headers={"User-Agent": "BreakPilot-Compliance-Agent/1.0"})
+            clean = re.sub(r"<[^>]+>", " ", resp.text)
+            return re.sub(r"\s+", " ", clean).strip()[:4000]
+    except Exception:
+        return ""
 
 
 async def _fetch_dse_html(url: str, scanned_pages: list[str]) -> str:
-    """Fetch the raw HTML of the privacy policy page (for structured parsing)."""
+    """Fetch the raw HTML of the privacy policy page."""
-    import re
+    dse_url = next((p for p in scanned_pages if re.search(r"datenschutz|privacy|dsgvo", p, re.IGNORECASE)), url)
-    dse_url = None
-    for page in scanned_pages:
-        if re.search(r"datenschutz|privacy|dsgvo", page, re.IGNORECASE):
-            dse_url = page
-            break
-    if not dse_url:
-        dse_url = url
     try:
         async with httpx.AsyncClient(timeout=15.0, follow_redirects=True) as client:
             resp = await client.get(dse_url, headers={"User-Agent": "BreakPilot-Compliance-Agent/1.0"})
@@ -541,7 +309,6 @@ def _build_findings(
             correction_text=ref.correction_text, insert_after=ref.insert_after,
         )
 
-    # Undocumented services (on website, NOT in DSE)
     for svc in comparison["undocumented"]:
         services.append(ServiceInfo(
             name=svc["name"], category=svc.get("category", "other"),
@@ -550,18 +317,16 @@ def _build_findings(
             requires_consent=svc.get("requires_consent", False),
             legal_ref=svc.get("legal_ref", ""), in_dse=False, status="undocumented",
         ))
-        severity = "HIGH" if is_live else "MEDIUM"
         ref = _get_ref(svc.get("id", ""))
         findings.append(ScanFinding(
             code=f"DSE-MISSING-{svc['id'].upper()}",
-            severity=severity,
+            severity="HIGH" if is_live else "MEDIUM",
             text=f"{svc['name']} ({svc.get('provider', '')}, {svc.get('country', '')}) "
                  f"ist auf der Website eingebunden aber NICHT in der Datenschutzerklaerung "
                  f"dokumentiert (Art. 13 DSGVO).",
             text_reference=ref,
         ))
 
-    # Documented services (OK)
     for item in comparison["documented"]:
         svc = item["detected"]
         services.append(ServiceInfo(
@@ -571,16 +336,13 @@ def _build_findings(
             requires_consent=svc.get("requires_consent", False),
             legal_ref=svc.get("legal_ref", ""), in_dse=True, status="ok",
         ))
-        # Check third-country transfer
         if not svc.get("eu_adequate", False):
             findings.append(ScanFinding(
-                code=f"TRANSFER-{svc['id'].upper()}",
+                code=f"TRANSFER-{svc['id'].upper()}", severity="MEDIUM",
-                severity="MEDIUM",
                 text=f"{svc['name']} ({svc.get('country', '')}) — Drittlandtransfer. "
                      f"Pruefen ob SCCs oder Angemessenheitsbeschluss dokumentiert sind.",
             ))
 
-    # Outdated services (in DSE, NOT on website)
     for svc in comparison["outdated"]:
         services.append(ServiceInfo(
             name=svc["name"], category="other",
@@ -595,15 +357,11 @@ def _build_findings(
                  f"nicht mehr gefunden. Eintrag bei naechster Aktualisierung entfernen.",
         ))
 
-    # Missing pages (e.g., /impressum returns 404)
     for page_url, status_code in scan.missing_pages.items():
         if "impressum" in page_url.lower():
             findings.append(ScanFinding(
-                code="MISSING-IMPRESSUM",
+                code="MISSING-IMPRESSUM", severity="HIGH",
-                severity="HIGH",
                 text=f"Impressum-Seite gibt HTTP {status_code} zurueck (§5 TMG Verstoss).",
             ))
 
     return services, findings
-
-