Hi MCP team!
Filing this as an informational nudge that a new community MCP server may be worth referencing in ADDITIONAL.md or a community list.
Django ORM Lens — a static-analysis MCP server that exposes Django ORM schemas to AI agents.
What it does
Five read-only tools any MCP client (Claude Desktop, Cursor, Aider, Continue.dev, Zed) can call while editing a Django project:
list_apps — every Django app in the workspace with model counts
list_models — flat app.Model list, optional app filter
describe_model — full field/relation/Meta detail for one model
find_relations — inbound + outbound relations for one model
er_diagram — Mermaid ER diagram for the whole workspace
Why it fits an MCP audience
- Static parser — no Django boot, no database, no credentials. Works offline, in CI, and on partial projects.
- Registered in the official MCP Registry as
io.github.FROWNINGdev/django-orm-lens.
- Listed on Glama.ai (safety checks passed) and mcp.so.
- Also ships as a companion VS Code extension and a zero-dep Python CLI — same parser, three surfaces.
- MIT licensed, PyPI:
pip install "django-orm-lens[mcp]".
Fully understand if this doesn't fit the reference-implementations scope of this repo — feel free to close if inappropriate. Thanks for the ecosystem!
Hi MCP team!
Filing this as an informational nudge that a new community MCP server may be worth referencing in ADDITIONAL.md or a community list.
Django ORM Lens — a static-analysis MCP server that exposes Django ORM schemas to AI agents.
What it does
Five read-only tools any MCP client (Claude Desktop, Cursor, Aider, Continue.dev, Zed) can call while editing a Django project:
list_apps— every Django app in the workspace with model countslist_models— flatapp.Modellist, optional app filterdescribe_model— full field/relation/Meta detail for one modelfind_relations— inbound + outbound relations for one modeler_diagram— Mermaid ER diagram for the whole workspaceWhy it fits an MCP audience
io.github.FROWNINGdev/django-orm-lens.pip install "django-orm-lens[mcp]".Fully understand if this doesn't fit the reference-implementations scope of this repo — feel free to close if inappropriate. Thanks for the ecosystem!