The Agent Skills Directory

[SAFE]: The skill is a set of documentation and tools for developers building MCP servers. All identified behaviors are consistent with its stated purpose of software development and testing.
[EXTERNAL_DOWNLOADS]: The instructions and scripts reference official Model Context Protocol documentation and SDKs hosted on modelcontextprotocol.io and GitHub. These are well-known resources for the technology the skill supports.
[COMMAND_EXECUTION]: The scripts/evaluation.py and scripts/connections.py files include functionality to launch and interact with local MCP servers using the stdio transport. This involves executing local commands as subprocesses, which is the standard and intended method for testing Model Context Protocol integrations locally.
[REMOTE_CODE_EXECUTION]: The evaluation harness utilizes the official anthropic and mcp libraries to communicate with AI models and execute tools provided by the MCP server under test. These operations are performed through standard API channels and are restricted to the context of the evaluation tasks provided by the user.

mcp-builder