llm-app-development

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill focuses entirely on defensive engineering for AI applications. It provides robust patterns for PII redaction, input length validation, and structured output verification using Zod, which are industry-standard security practices.
  • [SAFE]: The provided TypeScript examples for RAG, function calling, and streaming are well-structured and include necessary safety constraints, such as capping tool-use loops (maxSteps) to prevent infinite execution and using AbortControllers for timeouts.
  • [SAFE]: The metadata and documentation are consistent with the skill's stated purpose of helping developers move from basic prompting to production-ready AI systems.
  • [SAFE]: While the skill includes instructions to check for companion skills by listing directory contents, this is a standard developer experience enhancement for this platform and does not involve exfiltration or unauthorized access to sensitive files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 01:04 PM
Security Audit — agent-trust-hub — llm-app-development