Open Source Ready

English | 简体中文

An agent skill for getting private or personal repositories ready for a professional open-source launch.

Open Source Ready is intentionally plan-first. It audits a repository, separates blocking release risks from polish work, and produces a staged P0/P1/P2 upgrade plan before making changes. It is built for safety-sensitive release work: secrets, git history rewrites, license clarity, community health files, CI, versioning, and security posture.

Key Features

• Release readiness audit: checks license, README, community files, CI, Dependabot, changelog, tags, .gitignore, tracked .env files, and obvious secret patterns.
• Safety-first workflow: treats secret rotation, history rewrites, public pushes, and branch protection changes as explicit confirmation points.
• Professional repo templates: includes bilingual README templates, contribution/security/conduct docs, issue forms, PR template, Dependabot, and CI recipes.
• Bilingual by default: uses README.md as the English primary document and README.zh-CN.md as the Simplified Chinese document.
• Agent-neutral design: works as a readable skill folder for local coding agents with filesystem and shell access.

Install

Copy this repository into the skills directory used by your agent runtime:

SKILLS_DIR=/path/to/your/agent/skills
mkdir -p "$SKILLS_DIR"
git clone https://github.com/edward-zyz/open-source-ready.git "$SKILLS_DIR/open-source-ready"

Then start a new agent session or reload skills if your runtime supports manual skill refresh.

Quick Start

Run a read-only assessment against any repository:

bash scripts/assess.sh /path/to/repo

Use it through an agent:

Use $open-source-ready to assess this repo and prepare it for open-source release.

Natural prompts should also work:

I want to make this repo public on GitHub. First assess the risks and give me a plan.

What the Skill Does

The skill follows a staged workflow:

1. Assess the repository with scripts/assess.sh.
2. Plan P0/P1/P2 work before editing files.
3. Gate destructive steps such as git history rewrites or branch protection changes behind explicit confirmation.
4. Fill only the gaps instead of overwriting existing project-specific docs.
5. Validate the result with the same checks used by this repository.

Included Resources

.
├── SKILL.md
├── agents/
│   └── runtime metadata
├── assets/
│   ├── README.md
│   ├── README.zh-CN.md
│   ├── CODE_OF_CONDUCT.md
│   ├── CONTRIBUTING.md
│   ├── SECURITY.md
│   ├── PULL_REQUEST_TEMPLATE.md
│   ├── dependabot.yml
│   └── ISSUE_TEMPLATE/
├── examples/
│   └── assessment-sample.md
├── references/
│   ├── checklist.md
│   ├── ci-recipes.md
│   ├── sources.jsonl
│   └── upgrade-guide.md
└── scripts/
    └── assess.sh

Example Output

See examples/assessment-sample.md for the shape of an assessment report and how to interpret P0/P1/P2 findings.

Agent Compatibility

This repository is intentionally plain files plus shell/Python validation. An agent can use it in three ways:

• Read SKILL.md directly for the current session.
• Install the whole folder into its local skills directory.
• Copy individual templates from assets/ into a target repository during an open-source readiness pass.

No hosted service or package registry is required.

Safety Model

Open Source Ready treats publication as a risk-management workflow:

• Do not make a private repository public before P0 cleanup is complete.
• Rotate or revoke exposed secrets before deleting them from history.
• Confirm history rewrites separately because they change commit hashes.
• Prefer a new clean repository when extracting a skill or package from a private monorepo.
• Keep existing project files unless the assessment shows they are missing or the user asks for replacement.

Validation

Run the local validation commands:

python3 .github/scripts/validate_skill.py .
bash -n scripts/assess.sh
bash scripts/assess.sh .

The GitHub Actions workflow runs the same core checks on pull requests and pushes to main.

Roadmap

• Optional Scorecard workflow template.
• Optional release automation recipe for tagged skill releases.
• More ecosystem-specific CI and Dependabot examples.

Author

• X: @Edwardzyzt

License

MIT