Archived the add-mcp-server proposal to archive/2025-11-20-add-mcp-server/ and created the mcp-integration capability spec with 7 requirements. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
1.6 KiB
1.6 KiB
Add MCP Server for Member Management
Why
AI assistants need a standardized way to interact with the Clubber GraphQL API. The Model Context Protocol (MCP) provides a standard interface for AI tools to access application functionality. Currently, AI assistants can only interact with the API through raw GraphQL queries, which requires the assistant to construct queries manually and handle responses. An MCP server will provide structured tools that abstract away GraphQL complexity and enable seamless member management through AI agents like Claude.
What Changes
- Add official MCP Python SDK (
mcp) as dependency - Create MCP server implementation in
src/mcp_server.py - Implement 4 core tools for member management:
list_members- Query all members with their detailsget_member- Retrieve a single member by IDcreate_member- Create a new member (only firstName required)update_member- Update an existing member's information
- Use HTTP client (httpx) to connect to GraphQL API
- Configure via environment variable (CLUBBER_API_URL)
- Update README with MCP server setup, configuration, and usage instructions
- Document Claude Code integration steps
Impact
- Affected specs: New capability
mcp-integration - Affected code:
- New file:
src/mcp_server.py(~150-200 lines) - Modified:
pyproject.toml(add mcp dependency) - Modified:
README.md(add MCP server documentation)
- New file:
- User impact: Enables AI assistants to manage society members through natural language
- No breaking changes: This is additive functionality