The Agent Skills Directory

[COMMAND_EXECUTION]: The evaluation script (scripts/evaluation.py) is designed to launch and manage local MCP server processes using the stdio transport. This involves executing user-provided commands (e.g., python my_server.py) to facilitate automated testing of the server's tool implementation.- [EXTERNAL_DOWNLOADS]: The implementation guides (SKILL.md) recommend fetching official protocol specifications and SDK documentation from well-known sources, including modelcontextprotocol.io and the official Model Context Protocol GitHub organizations. These are standard resources for developers working with this protocol.- [INDIRECT_PROMPT_INJECTION]: The scripts/evaluation.py harness reads test questions from an XML file and interpolates them into a system prompt for the agent. This is an intended design feature for a testing framework where the developer provides their own test cases to evaluate server behavior.

protocol-implementation-framework