The Agent Skills Directory

[COMMAND_EXECUTION]: The connection utility (scripts/connections.py) and the evaluation harness (scripts/evaluation.py) allow the execution of arbitrary shell commands. These scripts are designed to start and manage MCP server processes locally via the stdio transport for development and testing purposes. The command and arguments are provided as user input through command-line flags (-c, -a).
[INDIRECT_PROMPT_INJECTION]: The evaluation harness (scripts/evaluation.py) reads task questions from an external XML file and interpolates them directly into a system prompt for the AI agent. While the script employs XML tags (<summary>, <feedback>, <response>) as boundary markers to guide the model, this pattern creates a surface for indirect prompt injection where instructions hidden in the evaluation data could influence agent behavior.
[EXTERNAL_DOWNLOADS]: The skill instructions guide the agent to fetch documentation from modelcontextprotocol.io and official GitHub repositories belonging to the modelcontextprotocol organization. These are well-known sources within the MCP ecosystem and are utilized to ensure implementation adheres to the latest protocol specifications.

mcp-builder