The Agent Skills Directory

[COMMAND_EXECUTION]: The evaluation harness in scripts/evaluation.py and the connection logic in scripts/connections.py enable the execution of local MCP servers through the stdio transport, using user-supplied command-line arguments.\n- [EXTERNAL_DOWNLOADS]: The documentation directs developers to fetch official protocol specifications and SDK details from the modelcontextprotocol GitHub organization, which is a recognized and authoritative source.\n- [PROMPT_INJECTION]: The utility in scripts/evaluation.py processes question-and-answer pairs from an external XML file for AI testing purposes. This creates a surface for indirect prompt injection if malicious data is provided in the evaluation set, although this risk is localized to the testing process.

mcp-builder