A TRAE Work Skill that automatically generates all the documentation files your open-source project needs (README, LICENSE, CONTRIBUTING, and 20+ more) by analyzing your project structure, code, and tech stack.
opensource-doc-generator is an AI Agent Skill that helps the AI agent generate a complete set of standardized documentation for open-source platforms (GitHub, GitLab, Gitee, etc.) based on your project's actual files and structure.
Core philosophy: Analyze the project first, then present tiered recommendation plans, and generate only after user confirmation.
Documentation quality directly impacts adoption rates, contributor growth, and community health. A good README lets users understand your project's value in 30 seconds. A clear CONTRIBUTING guide can cut contributor onboarding costs by 50%. And without a LICENSE file, your project isn't legally open source at all.
- Intelligent Project Analysis — Automatically scans project structure, config files, and tech stack to extract project name, version, dependencies, and more
- 20+ Document Types — Covers README, LICENSE, CONTRIBUTING, CHANGELOG, CODE_OF_CONDUCT, SECURITY, Issue templates, PR templates, and all common open-source docs
- Tiered Recommendation Plans — Generates 3-4 plans at different completeness levels (Full / Standard / Basic / Minimal) tailored to your project, letting you choose
- User Confirmation Mechanism — Before generating anything, presents a project summary, tiered plans, and questions (license type, doc language, author info) for you to confirm
- Decision Matrix Driven — Recommends documents based on project characteristics (accepts contributions, has releases, handles sensitive data, etc.)
- Tech Stack Aware — Recognizes Node.js / Python / Go / Rust / Java / C# / Godot and generates appropriate .gitignore rules
- License Selection Guide — Built-in guidance for MIT / Apache-2.0 / GPL-3.0 / BSD / LGPL / AGPL with use-case descriptions
- Detailed Content Guide — Each document type has a complete content structure template and best practices (see
references/file-types-guide.en.md) - Quality Checklist — Post-generation checks for completeness, cross-reference consistency, unfilled placeholders, and more
Clone this repo and copy the Skill files to your TRAE Work Skills directory:
git clone https://github.com/zyphraxns/opensource-doc-generator.git
cp opensource-doc-generator/SKILL.md ~/.trae-cn/skills/opensource-doc-generator/
cp -r opensource-doc-generator/references ~/.trae-cn/skills/opensource-doc-generator/- Download the ZIP of this repo
- Copy
SKILL.mdand thereferences/folder into~/.trae-cn/skills/opensource-doc-generator/
The Skill will auto-load in your next TRAE Work session.
After installation, just trigger it with natural language in TRAE Work:
Help me generate open-source docs for my project
Or more specifically:
I want to upload my project to GitHub, help me generate a README and other needed docs
The Skill will analyze your project, present recommendation plans for you to choose from, and generate the files after confirmation.
opensource-doc-generator/
├── SKILL.md # Skill main file (V2, English, default)
├── SKILL.zh.md # Skill main file (V2, Chinese)
├── references/
│ ├── file-types-guide.md # Detailed file type guide (Chinese)
│ └── file-types-guide.en.md # Detailed file type guide (English)
├── V1/ # First version (archived)
│ ├── SKILL.md
│ └── references/
│ └── file-types-guide.md
├── README.md # This file (English, default)
├── README.zh.md # Chinese README
├── LICENSE # MIT License
├── CONTRIBUTING.md # Contribution guide
├── CODE_OF_CONDUCT.md # Code of conduct
├── CHANGELOG.md # Changelog (bilingual)
├── .gitignore # Git ignore rules
├── .editorconfig # Editor config
├── .gitattributes # Git attributes
└── .github/
├── PULL_REQUEST_TEMPLATE.md # PR template
└── ISSUE_TEMPLATE/
├── bug_report.md # Bug report template
├── feature_request.md # Feature request template
└── config.yml # Issue template config
| Category | File | Description |
|---|---|---|
| Required | README.md |
Project front door, first entry point |
| Required | LICENSE |
Open-source license |
| Required | .gitignore |
Git ignore rules |
| Strongly Recommended | CONTRIBUTING.md |
Contribution guide |
| Strongly Recommended | CODE_OF_CONDUCT.md |
Code of conduct |
| Strongly Recommended | CHANGELOG.md |
Changelog |
| Community Health | SECURITY.md |
Security policy |
| Community Health | SUPPORT.md |
Support resources |
| Community Health | FUNDING.yml |
Sponsor config |
| Community Health | GOVERNANCE.md |
Project governance |
| Community Health | ISSUE_TEMPLATE/ |
Issue templates |
| Community Health | PULL_REQUEST_TEMPLATE.md |
PR template |
| Configuration | .editorconfig |
Editor style unification |
| Configuration | .gitattributes |
Git file attributes |
| Configuration | CODEOWNERS |
Code owners |
| Configuration | dependabot.yml |
Dependency auto-update |
| Scenario | CITATION.cff |
Academic citation |
| Scenario | NOTICE |
Attribution notice |
| Scenario | ARCHITECTURE.md |
Architecture doc |
| Scenario | ROADMAP.md |
Roadmap |
| Scenario | FAQ.md |
FAQ |
| Scenario | INSTALL.md |
Install guide |
| Scenario | Dockerfile |
Docker deployment |
| Scenario | docker-compose.yml |
Docker orchestration |
| Scenario | .env.example |
Env var example |
| Scenario | Makefile |
Build automation |
The Skill executes five steps:
- Project Analysis — Scans config files, tech stack, directory structure, existing docs
- Intelligent Tiered Recommendations — Internally generates 3-4 recommendation plans at different completeness levels
- User Confirmation & Selection — Presents project summary, tiered plans, and questions (license, language, author) — waits for your choice
- Document Generation — Generates files based on your confirmed plan
- Summary Report — Lists created files, suggestions, and files not created with reasons
| Version | Description |
|---|---|
| V2 (current) | Adds user confirmation step: presents tiered recommendation plans before generating |
V1 (archived in V1/) |
Original: analyzes and generates directly without user confirmation |
This project is licensed under the MIT License.