agent-builder-pydantic-ai
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS] (SAFE): The skill recommends installing standard, well-known Python packages (
pydantic-ai,httpx,pydantic,python-dotenv) from PyPI. No untrusted sources or piped execution patterns are present. - [CREDENTIALS_UNSAFE] (SAFE): The code demonstrates the correct use of
os.getenvfor managing sensitive API keys. The example environment variables use generic placeholders (e.g.,sk-or-v1-...) rather than hardcoded credentials. - [COMMAND_EXECUTION] (SAFE): There are no instances of dangerous system commands or arbitrary shell executions. The provided
pip installcommand is the standard method for managing Python dependencies. - [DATA_EXFILTRATION] (SAFE): No unauthorized network requests, sensitive file access (like SSH keys or AWS credentials), or data exfiltration patterns were identified.
- [PROMPT_INJECTION] (SAFE): System prompts used in the examples are generic and instructional, with no attempts to bypass safety filters or override agent behavior.
Audit Metadata