GlmMCP: Instructions file for Claude Code

GlmMCP

MCP (Model Context Protocol) server for GLM chat completions via AceDataCloud API.

Project Structure

core/
  config.py     — Settings dataclass (API token, base URL)
  server.py     — FastMCP server singleton
  client.py     — httpx async HTTP client
  types.py      — Literal types for model enum
  exceptions.py — Error classes (AuthError, APIError, TimeoutError)
tools/
  chat_tools.py  — glm_chat_completions
  info_tools.py  — glm_list_models, glm_get_usage_guide
prompts/           — LLM guidance prompts
tests/             — pytest-asyncio + respx tests

Sync from Docs

When invoked by the sync workflow, the Docs repo is checked out at _docs/. Your job:

1. Source of truth — _docs/openapi/glm.json is the OpenAPI spec for the GLM API.
2. Compare models — The Literal types in core/types.py must match the spec's model enum. Add/remove as needed.
3. Compare parameters — Each @mcp.tool() function's parameters should match the corresponding OpenAPI endpoint.
4. Update defaults — If a new model becomes the recommended default, update the default in core/types.py.
5. Update README — Keep the model table and feature list current.
6. Add tests — For new tools or parameters, add test cases in tests/.
7. PR title — Use format: sync: <description> [auto-sync]

Development

pip install -e ".[dev]"
pytest --cov=core --cov=tools
ruff check .