ai-integration-generator
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for generating AI features such as streaming chat, completions, and structured output that ingest untrusted user data and pass it directly to large language models.
- Ingestion points: The templates extract untrusted data from the
messagesandpromptfields of the incoming request body (req.json()) in generated API routes (SKILL.mdandassets/chat-route/route.ts). - Boundary markers: The generated code patterns lack delimiters or system instructions to ignore embedded commands, leaving the model vulnerable to instructions contained within the user input.
- Capability inventory: The generated code utilizes the Vercel AI SDK to perform Large Language Model calls via
streamText,generateObject, andembedfunctions. - Sanitization: The skill's templates do not include validation, filtering, or escaping logic to sanitize external content before interpolation into prompts.
Audit Metadata