The Agent Skills Directory

[SAFE]: The skill acts as a developer resource and testing harness. All external references target official repositories for the Model Context Protocol organization or well-known development services (GitHub).
[COMMAND_EXECUTION]: The evaluation script (scripts/evaluation.py) includes functionality to launch local processes via the stdio transport. This is a standard and documented feature of the Model Context Protocol for testing local server implementations and is gated by user-provided command-line arguments.
[EXTERNAL_DOWNLOADS]: The documentation encourages fetching official SDK documentation from the modelcontextprotocol GitHub organization. These are trusted sources and integral to the skill's purpose of teaching MCP development.
[DATA_EXFILTRATION]: The skill implements standard MCP client logic. While it supports network-based transports (HTTP/SSE), these are used for the primary purpose of interacting with the user's own MCP servers during development and testing.

mcp-builder