feat: Investor Agent — FAQ als LLM-Kontext statt Direkt-Streaming

Architektur-Umbau: FAQ-Antworten werden NICHT mehr direkt gestreamt.
Stattdessen werden die Top-3 relevanten FAQ-Einträge als Kontext
ans LLM übergeben. Das LLM interpretiert die Frage, kombiniert
mehrere FAQs bei komplexen Fragen und antwortet natürlich.

Vorher: Frage → Keyword-Match → FAQ direkt streamen (LLM umgangen)
Nachher: Frage → Top-3 FAQ-Matches → LLM-Prompt als Kontext → LLM antwortet

Neue Funktionen:
- matchFAQMultiple(): Top-N Matches statt nur bester
- buildFAQContext(): Baut Kontext-String für LLM-Injection
- faqContext statt faqAnswer im Request-Body
- System-Prompt Anweisung: "Kombiniere bei Bedarf, natürlicher Fließtext"

Behebt: Komplexe Fragen mit 2+ Themen werden jetzt korrekt beantwortet

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

pitch-deck/app/api/chat/route.ts

@@ -157,42 +157,12 @@ ${JSON.stringify(features.rows, null, 2)}
 export async function POST(request: NextRequest) {
   try {
     const body = await request.json()
-    const { message, history = [], lang = 'de', slideContext, faqAnswer } = body
+    const { message, history = [], lang = 'de', slideContext, faqContext } = body
 
     if (!message || typeof message !== 'string') {
       return NextResponse.json({ error: 'Message is required' }, { status: 400 })
     }
 
-    // FAQ shortcut: if client sends a pre-cached FAQ answer, stream it directly (no LLM call)
-    if (faqAnswer && typeof faqAnswer === 'string') {
-      const encoder = new TextEncoder()
-      const stream = new ReadableStream({
-        start(controller) {
-          // Stream the FAQ answer in chunks for consistent UX
-          const words = faqAnswer.split(' ')
-          let i = 0
-          const interval = setInterval(() => {
-            if (i < words.length) {
-              const chunk = (i === 0 ? '' : ' ') + words[i]
-              controller.enqueue(encoder.encode(chunk))
-              i++
-            } else {
-              clearInterval(interval)
-              controller.close()
-            }
-          }, 30)
-        },
-      })
-
-      return new NextResponse(stream, {
-        headers: {
-          'Content-Type': 'text/plain; charset=utf-8',
-          'Cache-Control': 'no-cache',
-          'Connection': 'keep-alive',
-        },
-      })
-    }
-
     const pitchContext = await loadPitchContext()
 
     let systemContent = SYSTEM_PROMPT
@@ -200,6 +170,11 @@ export async function POST(request: NextRequest) {
       systemContent += '\n' + pitchContext
     }
 
+    // FAQ context: relevant pre-researched answers as basis for the LLM
+    if (faqContext && typeof faqContext === 'string') {
+      systemContent += '\n' + faqContext
+    }
+
     // Slide context for contextual awareness
     if (slideContext) {
       const visited: number[] = slideContext.visitedSlides || []

pitch-deck/components/ChatFAB.tsx

@@ -7,7 +7,7 @@ import { ChatMessage, Language, SlideId } from '@/lib/types'
 import { t } from '@/lib/i18n'
 import { SLIDE_ORDER } from '@/lib/hooks/useSlideNavigation'
 import { PresenterState } from '@/lib/presenter/types'
-import { matchFAQ, getFAQAnswer } from '@/lib/presenter/faq-matcher'
+import { matchFAQMultiple, buildFAQContext } from '@/lib/presenter/faq-matcher'
 
 interface ChatFABProps {
   lang: Language
@@ -227,8 +227,9 @@ export default function ChatFAB({
     setIsStreaming(true)
     setIsWaiting(true)
 
-    // Check FAQ first for instant response
-    const faqMatch = matchFAQ(message, lang)
+    // Find relevant FAQ entries as context for the LLM
+    const faqMatches = matchFAQMultiple(message, lang, 3)
+    const faqContext = buildFAQContext(faqMatches, lang)
 
     abortRef.current = new AbortController()
 
@@ -245,9 +246,9 @@ export default function ChatFAB({
         },
       }
 
-      // If FAQ matched, send the cached answer for fast streaming (no LLM call)
-      if (faqMatch) {
-        requestBody.faqAnswer = getFAQAnswer(faqMatch, lang)
+      // Send FAQ context to LLM (not direct streaming — LLM interprets and combines)
+      if (faqContext) {
+        requestBody.faqContext = faqContext
       }
 
       const res = await fetch('/api/chat', {

pitch-deck/lib/presenter/faq-matcher.ts

@@ -7,11 +7,19 @@ import { PRESENTER_FAQ } from './presenter-faq'
  * Returns the best match if score exceeds threshold, or null for LLM fallback.
  */
 export function matchFAQ(query: string, lang: Language): FAQEntry | null {
+  const matches = matchFAQMultiple(query, lang, 1)
+  return matches.length > 0 ? matches[0] : null
+}
+
+/**
+ * Match a user query and return the top N relevant FAQ entries as context.
+ * Used to feed multiple relevant FAQs into the LLM prompt.
+ */
+export function matchFAQMultiple(query: string, lang: Language, maxResults: number = 3): FAQEntry[] {
   const normalized = query.toLowerCase().trim()
   const queryWords = normalized.split(/\s+/)
 
-  let bestMatch: FAQEntry | null = null
-  let bestScore = 0
+  const scored: { entry: FAQEntry; score: number }[] = []
 
   for (const entry of PRESENTER_FAQ) {
     let score = 0
@@ -20,23 +28,20 @@ export function matchFAQ(query: string, lang: Language): FAQEntry | null {
     for (const keyword of entry.keywords) {
       const kwLower = keyword.toLowerCase()
       if (kwLower.includes(' ')) {
-        // Multi-word keyword: check if phrase appears in query
         if (normalized.includes(kwLower)) {
           score += 3 * entry.priority / 10
         }
       } else {
-        // Single keyword: check word-level match
         if (queryWords.some(w => w === kwLower || w.startsWith(kwLower) || kwLower.startsWith(w))) {
           score += 1
         }
-        // Also check if keyword appears anywhere in query (partial match)
         if (normalized.includes(kwLower)) {
           score += 0.5
         }
       }
     }
 
-    // Check if query matches the question text closely
+    // Check question text overlap
     const questionText = lang === 'de' ? entry.question_de : entry.question_en
     const questionWords = questionText.toLowerCase().split(/\s+/)
     const overlapCount = queryWords.filter(w =>
@@ -46,22 +51,16 @@ export function matchFAQ(query: string, lang: Language): FAQEntry | null {
       score += overlapCount * 0.5
     }
 
-    // Weight by priority
     score *= (entry.priority / 10)
 
-    if (score > bestScore) {
-      bestScore = score
-      bestMatch = entry
+    if (score >= 1.0) {
+      scored.push({ entry, score })
     }
   }
 
-  // Threshold: need meaningful match to avoid false positives
-  // Require at least 2 keyword hits or strong phrase match
-  if (bestScore < 1.5) {
-    return null
-  }
-
-  return bestMatch
+  // Sort by score descending, return top N
+  scored.sort((a, b) => b.score - a.score)
+  return scored.slice(0, maxResults).map(s => s.entry)
 }
 
 /**
@@ -70,3 +69,18 @@ export function matchFAQ(query: string, lang: Language): FAQEntry | null {
 export function getFAQAnswer(entry: FAQEntry, lang: Language): string {
   return lang === 'de' ? entry.answer_de : entry.answer_en
 }
+
+/**
+ * Build a context string from multiple FAQ matches for LLM injection
+ */
+export function buildFAQContext(entries: FAQEntry[], lang: Language): string {
+  if (entries.length === 0) return ''
+
+  const parts = entries.map((entry, idx) => {
+    const q = lang === 'de' ? entry.question_de : entry.question_en
+    const a = lang === 'de' ? entry.answer_de : entry.answer_en
+    return `### Relevante Information ${idx + 1}: ${q}\n${a}`
+  })
+
+  return `\n\n## Vorrecherchierte Antworten (nutze diese als Basis, kombiniere bei Bedarf)\n${parts.join('\n\n')}\n\nWICHTIG: Formuliere die Antwort in deinen eigenen Worten als natürlichen Fließtext. Kombiniere die Informationen wenn die Frage mehrere Themen berührt. Antworte nicht mit Bulletlisten.`
+}