agentleFS
Sign inStart free
NVIDIAKnown publisherNVIDIA/auto-ontology/CLAUDE.md

auto-ontology

CLAUDE.md9.9k starsChanged 43 days agoApache-2.0View on GitHub

Full-stack repo: Next.js frontend (/frontend) + FastAPI backend (/auto_ontology). This file is for Auto-Ontology maintainers. User/partner agent skills live in skills/ — start at AGENTS.md. Backend commands run from the repo root: The app factory is create_app() in auto_ontology/server/__main__.py — there is no auto_ontology/server/main.py. In frontend/CLAUDE.md, loaded when working under frontend/. A PR is not done until every GitHub Actions workflow passes. After opening one, poll gh pr checks <number> until no check is pending, then fix whatever failed and…

Trending

More talk lately.about the whole repository

Dec 2025: 0 mentionsJan 2026: 0 mentionsFeb 2026: 0 mentionsMar 2026: 0 mentionsApr 2026: 0 mentionsMay 2026: 0 mentionsJun 2026: 0 mentionsJul 2026: 0 mentionsAug 2026: 0 mentionsSep 2026: 0 mentionsOct 2026: 0 mentionsNov 2026: 64 mentions
Dec 2025Jun 2026Nov 2026
Mentions a month of the NVIDIA/auto-ontology repository that our weekly crawl found. There is no adoption data for this file yet.

Who's talking about NVIDIA/auto-ontology

The biggest 5 of 64 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 64

Talked about on GitHub, YouTube, Bluesky and Hacker News (64 times).

On GitHub · 27

22 more on GitHub

On YouTube · 24

19 more on YouTube

On Bluesky · 12

7 more on Bluesky

On Hacker News · 1

The file

We split this CLAUDE.md into the rules it gives an agent

6 rules. About 0 are common, 1 is distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Formatting/Linting: Run uv run ruff check auto_ontology/ and uv run ruff format auto_ontology/ from repo root before committing.

    line 34
  2. Very specific:

    API prefix: All routes under /api/. Routes live in auto_ontology/server/datasources/router.py, orchestration in …

    line 36
  3. Very specific:

    Exception handlers: Use plain def (not async def) for @app.exception_handler functions.

    line 38
  4. Very specific:

    Commit messages: concise imperative style, no Co-Authored-By trailers.

    line 42
  5. Very specific:

    A PR is not done until every GitHub Actions workflow passes: After opening one, poll gh pr checks <number> until no check is pending, then …

    line 47
  6. Distinctive:

    Run these locally first — they are the same gates CI applies, and catching a failure here costs seconds instead of a CI round-trip:

    line 52

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md
<!-- SPDX-FileCopyrightText: Copyright (c) 2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved. SPDX-License-Identifier: Apache-2.0 -->

Auto-Ontology

Full-stack repo: Next.js frontend (/frontend) + FastAPI backend (/auto_ontology).

This file is for Auto-Ontology maintainers. User/partner agent skills live in skills/ — start at AGENTS.md.

Dev Commands

Backend commands run from the repo root:

CommandDescription
uv run uvicorn auto_ontology.server.__main__:create_app --factory --reload --host 127.0.0.1 --port 3001Start FastAPI on :3001 with hot reload
uv run python -m auto_ontology.serverStart FastAPI on :3001 without reload (what the container runs)
uv run python -m auto_ontology.ingestion_serviceStart the ingestion service on :3002

The app factory is create_app() in auto_ontology/server/__main__.py — there is no auto_ontology/server/main.py.

Frontend Rules

In frontend/CLAUDE.md, loaded when working under frontend/.

Backend Rules

  • The store is Postgres. auto_ontology/dal/ queries it with SQLAlchemy Core (no ORM); auto_ontology/dal/schema.py is the single source of truth for the schema and auto_ontology/dal/session.py owns the pooled engine. Schema changes go through Alembic — uv run alembic revision --autogenerate -m "..." — never by hand-editing a migration that has been applied. Several column types and constraints are deliberate and non-obvious — read the comments in schema.py before changing a table.
  • Formatting/Linting: Run uv run ruff check auto_ontology/ and uv run ruff format auto_ontology/ from repo root before committing.
  • Dependencies: Managed with uv. Add dependencies via uv add, not pip. Do not edit pyproject.toml manually for deps.
  • API prefix: All routes under /api/. Routes live in auto_ontology/server/datasources/router.py, orchestration in auto_ontology/server/datasources/service.py.
  • Type hints: Required on all function signatures.
  • Exception handlers: Use plain def (not async def) for @app.exception_handler functions. They perform no async I/O, so synchronous handlers are preferred.

Git

  • Commit messages: concise imperative style, no Co-Authored-By trailers.
  • Branch from main.

Opening a PR

A PR is not done until every GitHub Actions workflow passes. After opening one, poll gh pr checks <number> until no check is pending, then fix whatever failed and push again — repeat until the run is green. Report the final state honestly; do not hand back a PR with unexamined or failing checks.

Run these locally first — they are the same gates CI applies, and catching a failure here costs seconds instead of a CI round-trip:

CheckCommandCatches
Ruffuv run ruff check auto_ontology/ && uv run ruff format --check auto_ontology/Backend lint/format
Pytestuv run pytest auto_ontology dev_tools -qBackend tests
OpenAPIuv run python -m dev_tools.generate_backend_openapiStale docs/openapi/*.json
Frontendcd frontend && npx prettier --check . && pnpm lint && pnpm buildFormat, lint, build

The OpenAPI one is the easy one to miss: docs/openapi/backend.json and ingestion.json are committed, and the spec embeds each route's docstring. Editing a docstring on any @router handler — not just changing a signature or model — makes the committed spec stale and fails "Backend spec is up to date". Regenerate and commit the result whenever a route's code or docs change.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free