ai-wrapper-product

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as an educational and structural template for developing AI products, focusing on integration with standard LLM APIs.
  • [EXTERNAL_DOWNLOADS]: The provided code examples reference established and well-known libraries, including the official @anthropic-ai/sdk and the utility package p-queue for request management.
  • [DATA_EXPOSURE]: The documentation explicitly includes validation checks that warn against exposing API keys and recommends secure handling via environment variables and backend logic.
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates how to interpolate user input into prompt templates. While this creates a potential surface for indirect injection, the skill acknowledges this risk by including 'Input Validation + Sanitization' in its architecture diagram and implementing basic input length constraints in code snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 08:36 AM
Security Audit — agent-trust-hub — ai-wrapper-product