mirror of
https://github.com/vinta/awesome-python.git
synced 2026-10-02 08:23:10 +08:00
AGENTS.md still carried pre-reform Key Rules (mandatory alphabetical ordering, "quality over quantity" lanes-era language, no prune-sweep commit exception, no live-verification rule). Bring it in line with CLAUDE.md and CONTRIBUTING.md: downloads-descending tier ordering, 3+2 cap, sweep-commit exception, verification rule. Also record the UV_PYTHON=3.13 workaround for Python-3.14-only machines in both files' Makefile notes, and complete CLAUDE.md's Makefile target list. Co-Authored-By: Claude <noreply@anthropic.com>
2.5 KiB
2.5 KiB
CLAUDE.md
Repository Overview
An opinionated list of Python frameworks, libraries, tools, and resources. Published at awesome-python.com.
Entry Guidelines
Refer to CONTRIBUTING.md for admission rules, quality requirements, rejection rules, and entry format. Apply these rules whenever adding or removing an entry, whether reviewing a PR or committing directly.
Structure
- README.md: Source of truth. Hierarchical categories; entries ordered per the Key Rules below.
- CONTRIBUTING.md: Submission guidelines and review criteria.
- SPONSORSHIP.md: Sponsor tiers, placement rules, and the editorial-independence policy. Sponsor content sits in the README header and must never influence which projects get listed.
- website/: Static site generator that builds awesome-python.com from README.md.
build.py: Parses README.md and renders HTML via Jinja2 templates.fetch_github_stars.py: Fetches star counts intowebsite/data/.readme_parser.py: Markdown-to-structured-data parser.templates/,static/: Jinja2 templates and CSS/JS assets.tests/: Pytest tests for the build pipeline.
- Makefile:
make install,make build,make preview,make test,make lint,make format,make typecheck,make fetch_github_stars. On machines with only Python 3.14, prefix uv-based targets withUV_PYTHON=3.13(watchdog 6.0.0 ships no 3.14 wheel and the project setsno-build). - pyproject.toml: Uses
uvfor dependency management. Python >=3.13.
Key Rules
- Ordering within a use case: obvious choices first, then challengers, each tier by PyPI downloads/month descending; no-signal entries last in tier, alphabetically. See CONTRIBUTING.md.
- A shortlist, not a catalog: per use case, up to 3 obvious choices plus up to 2 challengers, hard maximum 5.
- One project per PR.
- One entry per commit when adding or deleting entries. Exception: a prune sweep is one commit per section, its body listing each removal with its reason. Format, wording, or categorization changes across multiple entries may be bundled in a single commit.
- Every keep/drop reason must be verified against current online data at decision time — download counts, repo activity and archived status, PyPI metadata, project docs. Training-data recollections alone are not evidence; verify before stating, and label anything unverifiable as a judgment call.
- README.md is the single source of content truth.