🛡️ Sentinel: [HIGH] Secure input validation for transcript and notes API

.jules/sentinel.md

@@ -0,0 +1,4 @@
+## 2025-05-23 - [Critical Security Learning - Sentinel Init]
+**Vulnerability:** Initial scan revealed minor issues but solid foundation.
+**Learning:** Project uses Supabase RLS and `withSecurity` middleware.
+**Prevention:** Maintain patterns.

app/api/notes/route.ts

@@ -1,11 +1,11 @@
 import { NextRequest, NextResponse } from 'next/server';
 import { createClient } from '@/lib/supabase/server';
 import { withSecurity, SECURITY_PRESETS } from '@/lib/security-middleware';
-import { formatValidationError, noteDeleteSchema, noteInsertSchema } from '@/lib/validation';
+import { formatValidationError, noteDeleteSchema, noteInsertSchema, youtubeIdSchema } from '@/lib/validation';
 import { z } from 'zod';
 
 const getNotesQuerySchema = z.object({
-  youtubeId: z.string()
+  youtubeId: youtubeIdSchema
 });
 
 interface NoteRow {

app/api/transcript/route.ts

@@ -4,6 +4,15 @@ import { withSecurity, SECURITY_PRESETS } from '@/lib/security-middleware';
 import { shouldUseMockData, getMockTranscript } from '@/lib/mock-data';
 import { mergeTranscriptSegmentsIntoSentences } from '@/lib/transcript-sentence-merger';
 import { NO_CREDITS_USED_MESSAGE } from '@/lib/no-credits-message';
+import { z } from 'zod';
+
+// Strict validation for language code (ISO 639-1)
+// Only allow 2-10 alphanumeric characters and dashes
+const langSchema = z.string()
+  .min(2)
+  .max(10)
+  .regex(/^[a-zA-Z0-9-]+$/)
+  .optional();
 
 function respondWithNoCredits(
   payload: Record<string, unknown>,
@@ -33,6 +42,13 @@ async function handler(request: NextRequest) {
       return respondWithNoCredits({ error: 'Invalid YouTube URL' }, 400);
     }
 
+    // Validate lang if provided
+    try {
+      langSchema.parse(lang);
+    } catch {
+      return respondWithNoCredits({ error: 'Invalid language code format' }, 400);
+    }
+
     if (shouldUseMockData()) {
       console.log(
         '[TRANSCRIPT] Using mock data (NEXT_PUBLIC_USE_MOCK_DATA=true)'
@@ -210,7 +226,7 @@ async function handler(request: NextRequest) {
     }
 
     const rawSegments = Array.isArray(transcriptSegments)
-      ? transcriptSegments.map((item, idx) => {
+      ? transcriptSegments.map((item) => {
         const transformed = {
           text: item.text || item.content || '',
           // Convert milliseconds to seconds for offset/start