ai-product
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill documentation contains strings like 'Ignore all previous instructions' and 'reveal your system prompt'. These are used within a 'Sharp Edges' section to educate developers on potential security vulnerabilities and demonstrate the importance of input sanitization.
- [SAFE]: The skill explicitly promotes defensive engineering practices, such as validating LLM outputs against schemas (using Zod), avoiding hardcoded API keys, and implementing circuit breakers for API failures.
- [EXTERNAL_DOWNLOADS]: The skill references industry-standard, well-known libraries and services including OpenAI, the Vercel AI SDK, Tiktoken, and Zod in its implementation examples.
Audit Metadata