mcp-python-conventions
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate technical guidelines for the Model Context Protocol (MCP), emphasizing the importance of routing logs to stderr to prevent corruption of the JSON-RPC stdout stream.
- [SAFE]: It advocates for security best practices in error handling by advising developers to use explicit ToolError exceptions to control the information exposed to LLM clients, rather than leaking internal details.
- [SAFE]: All external resources and URLs referenced in the rationale and code snippets point to official documentation and reputable primary sources (Anthropic, MCP specification).
- [SAFE]: The provided Python code snippet is a functional, safe example that demonstrates the described conventions using established libraries like fastmcp and pydantic.
- [SAFE]: No patterns of obfuscation, credential theft, or unauthorized command execution were detected.
Audit Metadata