henryperkins

henryperkins/my-mcp-github

Agent Claude Code

Use this agent when you need to evaluate the quality, safety, and effectiveness of available tools (both MCP and built-in tools). This includes assessing tool clarity, input schema design, testability, and real-world performance. The agent should be invoked after tools are created or modified, during tool selection…

henryperkins/my-mcp-github

Hook Claude Code

Runs when you submit a prompt, before the agent sees it, executing planning-prompt.py. From henryperkins/my-mcp-github.

henryperkins/my-mcp-github

Settings file Claude Code

Agent settings declaring 1 hook event (UserPromptSubmit).

henryperkins/my-mcp-github

Instructions file

Claude Code instructions for henryperkins/my-mcp-github, covering claude.md, project overview, architecture, core components and key implementation details.