Audit, polish, and package a repository before you open-source it, share it, or submit it to community directories.
For Chinese docs, see README.zh-CN.md.
For the skill definition, see SKILL.md.
github-repo-commander is a general-purpose repository stewardship skill. It combines public-facing README polish, privacy and metadata checks, and deeper GitHub workflow guidance so a repo becomes safer to publish and easier to recommend.
- maintainers preparing a repo for public release
- skill or tool authors improving README, metadata, and discoverability
- people cleaning privacy leaks, weak packaging, and stale docs in one pass
- community contributors preparing repos for awesome lists or curated directories
- catch privacy and open-source blockers before publishing
- improve README first screen and project positioning
- align metadata across
README,SKILL.md,skill.json, and package manifests - strengthen trust signals such as changelogs, security policy, and contribution guidance
- prepare a repo for community submission or directory review
Clone the repo into your skills directory, or attach SKILL.md directly in ecosystems that support file-based skills.
git clone https://github.com/wd041216-bit/github-repo-commander.git \
~/.codex/skills/github-repo-commanderRecommended prerequisites:
ghgit- Python 3 for the audit helper
python3 ./scripts/repo_commander_audit.py /path/to/repoTypical findings:
HIGH Potential github_token exposed in README.md
HIGH Local machine path found in docs/setup.md
MEDIUM README.md is missing a clear public-facing first screen
LOW CHANGELOG.md is missing for a public package-style repo
LOW No SECURITY.md found for a repo that advertises security checks
See examples/audit-example.md for a fuller sample.
- Audit the repo for privacy, packaging, and metadata gaps.
- Fix critical blockers first:
- tokens
- local paths
- personal data
- misleading public docs
- Improve the first screen:
- headline
- audience
- quick start
- examples
- Align packaging and trust signals:
- description
- topics
- changelog
- contributing docs
- security policy
- Prepare for community submission:
- awesome-list entry
- alphabetical placement
- metadata consistency
Most repo tools focus on one slice of the problem: secret scanning, README generation, or discoverability. github-repo-commander treats GitHub readiness as one workflow:
- make it safe
- make it clear
- make it credible
- make it discoverable
That makes it a better fit for real repository cleanup, where presentation, policy, and metadata usually need to move together.
This repo still includes deeper workflow material for heavier GitHub operations:
- references/workflow.md: extended workflow guide
- references/gh-commands.md: GitHub CLI reference
- scripts/repo-audit.sh: legacy shell audit helper
- scripts/privacy-check.sh: privacy-focused checks
- scripts/competitor-search.sh: competitor repo search helper
This repo is intentionally packaged as a public community skill:
- use SKILL.md as the main instruction surface
- use skill.json and _meta.json for ecosystem metadata
- use agents/openai.yaml for skill-card metadata
- use scripts/repo_commander_audit.py when you want a deterministic repo audit helper
- SKILL.md: main skill instructions
- README.zh-CN.md: Chinese overview
- examples/audit-example.md: sample audit findings
- docs/compatibility.md: compatibility and portability notes
- CONTRIBUTING.md: contribution workflow
- SECURITY.md: vulnerability reporting policy
- CHANGELOG.md: notable upgrades
- scripts/repo_commander_audit.py: repo audit helper
MIT