chore(search): diagnose unsupported Fireflies response shapes (#9488)

* test(search): guard Fireflies response diagnostic privacy

* chore(search): diagnose unsupported Fireflies response shapes

* test(search): cover encoded and oversized diagnostic text

* fix(search): retain bounded provider diagnostics in production

Author: waleedlatif1

apps/sim/lib/sim-search/live/fireflies-mcp.ts

@@ -1,3 +1,4 @@
+import { createLogger } from '@sim/logger'
 import { toArray, toRecord } from '@sim/utils/object'
 import { nativeText } from '@/lib/sim-search/live/dates'
 import { NativeSearchError, string } from '@/lib/sim-search/live/http'
@@ -6,7 +7,49 @@ import { boundedMeetingContent } from '@/lib/sim-search/live/meeting-content'
 import type { NativeDocument, NativePage, NativeSearchInput } from '@/lib/sim-search/live/types'
 
 const MAX_MEETING_CONTENT = 200_000
+const MAX_DIAGNOSTIC_TEXT_LENGTH = 65_536
 const UTC_DAY_MS = 86_400_000
+const logger = createLogger('FirefliesMcpSearch', { logLevel: 'WARN' })
+
+/** Only fixed envelope keys and value kinds may enter diagnostics, never provider keys or values. */
+function responseShape(value: unknown) {
+  const record = toRecord(value)
+  const kind = (entry: unknown) =>
+    entry === null ? 'null' : Array.isArray(entry) ? 'array' : typeof entry
+  return {
+    type: kind(value),
+    fields: Object.fromEntries(
+      [
+        'transcripts',
+        'data',
+        'result',
+        'results',
+        'items',
+        'output',
+        'text',
+        'error',
+        'errors',
+      ].flatMap((key) => (Object.hasOwn(record, key) ? [[key, kind(record[key])]] : []))
+    ),
+  }
+}
+
+function responseTextShape(value: unknown) {
+  if (typeof value !== 'string') return undefined
+  if (value.length > MAX_DIAGNOSTIC_TEXT_LENGTH)
+    return { length: value.length, inspection: 'skipped_size_limit' }
+  const text = value.trim()
+  const fenced = text.match(/^```(?:json)?\r?\n([\s\S]*)\r?\n```$/i)?.[1]
+  let jsonShape: ReturnType<typeof responseShape> | undefined
+  try {
+    jsonShape = responseShape(JSON.parse(fenced ?? text))
+  } catch {}
+  return {
+    length: value.length,
+    format: fenced !== undefined ? 'code_fence' : jsonShape ? 'json' : 'text',
+    ...(jsonShape ? { jsonShape } : {}),
+  }
+}
 
 function requireMeetingId(id: string): string {
   if (!/^[\w-]{1,200}$/.test(id))
@@ -139,11 +182,19 @@ export async function searchFirefliesMcp(
   const rows = Array.isArray(result)
     ? result
     : (data.transcripts ?? toRecord(data.data).transcripts)
-  if (!Array.isArray(rows))
+  if (!Array.isArray(rows)) {
+    logger.warn('Unsupported Fireflies transcript list response shape', {
+      response: responseShape(result),
+      data: responseShape(data.data),
+      result: responseShape(data.result),
+      output: responseShape(data.output),
+      text: responseTextShape(typeof result === 'string' ? result : data.text),
+    })
     throw new NativeSearchError(
       'unavailable',
       'Fireflies returned an unsupported transcript list format.'
     )
+  }
   const documents = rows.slice(0, limit).flatMap((value) => {
     const document = meetingDocument(toRecord(value))
     return document ? [document] : []

apps/sim/lib/sim-search/live/meeting-mcp.test.ts

@@ -1,3 +1,4 @@
+import { getAllMockLoggers } from '@sim/testing/mocks/logger.mock'
 import { describe, expect, it } from 'vitest'
 import { readFirefliesMcp, searchFirefliesMcp } from '@/lib/sim-search/live/fireflies-mcp'
 import { readGranolaMcp, searchGranolaMcp } from '@/lib/sim-search/live/granola-mcp'
@@ -157,6 +158,60 @@ describe('meeting MCP provider wire contracts', () => {
     await expect(searchFirefliesMcp(client, input)).rejects.toThrow('unsupported')
   })
 
+  it.each(['text', 'json', 'code_fence', 'oversized'] as const)(
+    'diagnoses unsupported Fireflies %s envelopes without logging private values or arbitrary keys',
+    async (format) => {
+      const privateMarker = 'PRIVATE_MEETING_OR_CREDENTIAL'
+      const json = JSON.stringify({ items: [{ title: privateMarker }], [privateMarker]: 'secret' })
+      const text =
+        format === 'oversized'
+          ? JSON.stringify({ items: [privateMarker.repeat(3000)] })
+          : format === 'code_fence'
+            ? ['```json', json, '```'].join('\n')
+            : format === 'json'
+              ? json
+              : privateMarker.repeat(1000)
+      const client: ManagedSearchMcpClient = {
+        async call() {
+          return {
+            text,
+            data: { error: privateMarker },
+            result: { items: [{ title: privateMarker }] },
+            ...Object.fromEntries(
+              Array.from({ length: 1000 }, (_, index) => [
+                `${privateMarker}_${index}`,
+                privateMarker,
+              ])
+            ),
+          }
+        },
+      }
+      await expect(searchFirefliesMcp(client, { ...input, query: privateMarker })).rejects.toThrow(
+        'unsupported'
+      )
+      const warnings = getAllMockLoggers().flatMap((logger) => logger.warn.mock.calls)
+      expect(warnings).toHaveLength(1)
+      expect(warnings[0]?.[1]).toMatchObject({
+        response: { type: 'object', fields: { text: 'string', data: 'object', result: 'object' } },
+        data: { type: 'object', fields: { error: 'string' } },
+        result: { type: 'object', fields: { items: 'array' } },
+        text:
+          format === 'oversized'
+            ? { length: text.length, inspection: 'skipped_size_limit' }
+            : {
+                length: text.length,
+                format,
+                ...(format === 'text'
+                  ? {}
+                  : { jsonShape: { type: 'object', fields: { items: 'array' } } }),
+              },
+      })
+      const serialized = JSON.stringify(warnings)
+      expect(serialized).not.toContain(privateMarker)
+      expect(serialized.length).toBeLessThan(2000)
+    }
+  )
+
   it('rejects invalid continuation offsets before Fireflies receives a query', async () => {
     const client = {
       async call() {