Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    Open Source Docs

    1

    Automate FSF-compliant open-source governance, licensing, security policies, and GitHub community health files.

    Free

    Security scanned
    Open Source Docs

    Open Source Docs

    Example session with this skill installed

    I want to open-source my React dashboard project. I need a license, readme, contributing guide, and security policy. I'm using the MIT license.

    • Read your context and instructions
    • Compiled the open source docs

    Conversation

    1. Survey

    Agent:

    Which parts of the open-source strategy do you want to implement? Reply with the numbers, a range, or "all".

    1. README.md
    2. LICENSE
    3. REUSE.toml
    4. CONTRIBUTING.md
    5. SECURITY.md
    6. CODE_OF_CONDUCT.md
    7. GOVERNANCE.md
    8. CHANGELOG.md
    9. Issue and pull request templates in .github/
    10. GitHub Actions workflows in .github/workflows/
    11. Dependabot in .github/
    12. Conventional commits
    13. Commit authentication with GPG + DCO
    14. Git flow and pull requests
    15. ARCHITECTURE_DECISIONS.md

    From your message I understand 1, 2, 4 and 5. Confirm or add more.

    User: 1, 2, 4, 5

    2. Step 0: what already exists

    Agent:

    These files already exist in the repository:

    • README.md (1.2 KB, first line: # React + Vite)

    What should I do with it?

    1. Overwrite: replace it with the generated one.
    2. Merge: keep what you wrote and add only what is missing.
    3. Skip: leave it untouched and generate nothing for it.
    4. Write .new: generate README.md.new next to it (default).

    LICENSE, CONTRIBUTING.md and SECURITY.md do not exist yet, so they will be created directly.

    User: merge

    3. License

    Agent:

    LicenseTypeIn one sentence
    MITPermissiveThe simplest and most popular one: do whatever you want with the code, just keep the copyright notice. Maximum adoption, no patent clause.
    Apache-2.0PermissiveLike MIT but with an explicit patent grant and protection against lawsuits; recommended for projects with patent implications or corporate backing.
    BSD-3-ClausePermissiveVery close to MIT, adding a clause that forbids using the authors' names to promote derivatives without permission.
    GPL-3.0CopyleftAny derivative work must be distributed under the same license (open source is mandatory). It protects the software's freedom, but reduces commercial adoption.
    AGPL-3.0Network copyleftGPL-3.0 extended to network services: if you offer the software as a web service, you must publish the source code to its users. Ideal for SaaS that wants to avoid closed appropriation.

    You said MIT. Do you confirm? Remember that all of these licenses are delivered with no warranties.

    User: Yes, MIT.

    4. Project data

    Agent:

    A few details before I write anything

    1. In which language should the documentation be generated? (English by default)
    2. What is the project name?
    3. How long will it take to review and resolve a contribution, and what is the maintainers' availability (for example, are they students)?
    4. Was the project built at a hackathon? If so, which one?
    5. Full name and email address of the maintainers?

    User: English. "Pulse Dashboard". About a week, I work on it in my spare time. No hackathon. Jane Doe jane@example.com

    5. Generation

    The agent works without further questions

    1. Reads only the references for the marked options: readme.md, contributing.md and security.md (license.md was already read in step 3).
    2. Downloads the MIT text from the canonical SPDX source into LICENSE, copies it to LICENSES/MIT.txt, and fills in the year (2026) and the copyright holder (Jane Doe). The license text is never typed by hand.
    3. Writes docs/COMPONENTS_LICENSE.md, explaining why MIT was chosen and how it fits the licenses of the project's dependencies (React and the rest of package.json).
    4. Merges README.md: keeps the existing content and adds the purpose, features, installation, usage examples, configuration, an architecture diagram (Mermaid), a supported platforms table, troubleshooting and support channels. The REUSE and OpenSSF badges are left commented out with sign-up instructions, because the project is not registered yet.
    5. Writes CONTRIBUTING.md: thanks to contributors, the one-week review expectation, environment setup and build steps, coding standards derived from the project's existing ESLint/Prettier configuration, tests, the pull request process and the definition of done.
    6. Writes SECURITY.md: supported versions, private reporting channel (GitHub Private Vulnerability Reporting or jane@example.com), response timelines, known limitations and coordinated disclosure policy.

    Result

    Files delivered

    pulse-dashboard/
    ├── README.md                    # merged with the existing one
    ├── LICENSE                      # MIT, downloaded from SPDX
    ├── CONTRIBUTING.md              # new
    ├── SECURITY.md                  # new
    ├── LICENSES/
    │   └── MIT.txt                  # copy of LICENSE
    └── docs/
        └── COMPONENTS_LICENSE.md    # rationale for choosing MIT
    

    Final summary shown to the user

    Agent:

    Done. Here is what changed:

    FileAction
    README.mdMerged with your existing file
    LICENSECreated (MIT)
    LICENSES/MIT.txtCreated
    docs/COMPONENTS_LICENSE.mdCreated
    CONTRIBUTING.mdCreated
    SECURITY.mdCreated

    Verification

    CheckResult
    Every marked option has its file on diskPassed
    No unsubstituted placeholdersPassed
    LICENSE and LICENSES/MIT.txt are over 500 bytes and start with MIT LicensePassed
    REUSE lintNot run (REUSE.toml was not selected)
    YAML syntaxNot run (no workflows or templates were selected)

    Next steps for you: enable Private Vulnerability Reporting in the repository's Security tab, and register the project on bestpractices.dev and scorecard.dev to activate the badges that are commented out in the README.

    1. README.md

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Open-source projects often lack the essential governance files required for community adoption, security compliance, and legal protection. This skill automates the creation of a complete open-source framework, ensuring your repository meets FSF and REUSE standards without manual drafting.

    What it does

    • Governance Generation produces critical community health files including CONTRIBUTING, CODE_OF_CONDUCT, and GOVERNANCE docs.
    • License Automation fetches OSI-approved licenses and sets up REUSE-compliant SPDX headers and LICENSES directories.
    • Security Compliance drafts SECURITY.md policies aligned with the EU Cyber Resilience Act (CRA).
    • CI/CD Scaffolding configures GitHub Actions for builds, linting, and security scanning, plus Dependabot for CVE tracking.
    • Release Management establishes CHANGELOG structures, conventional commit standards, and git flow procedures.

    How it works

    1. Survey Selection allows you to choose from 15 specific open-source artifacts via a numbered list.
    2. Conflict Resolution scans your current directory for existing files to determine if it should merge, overwrite, or skip.
    3. Data Collection gathers project-specific metadata like maintainer contacts and hackathon details only for the selected modules.
    4. Validation Loop runs automated checks including YAML syntax parsing and reuse lint to ensure all files are valid and placeholders are replaced.

    Frameworks & tools

    The skill utilizes Python for YAML validation, pipx for REUSE linting, and the GitHub CLI for API interactions. It supports various OSI-approved licenses and adheres to Keep a Changelog and Contributor Covenant standards.

    Why this beats prompting it yourself

    Generic prompts often miss the strict typographic and legal requirements of SPDX headers or the specific security mandates of the EU CRA. This skill executes a systematic validation loop that prevents broken YAML or empty license files from reaching your main branch.

    Use cases

    • Transitioning a private internal tool to a public open-source repository.
    • Preparing a hackathon project for submission with proper attribution and licensing.
    • Standardizing community health files across an entire GitHub organization.
    • Retrofitting a legacy repository to meet modern FSF and REUSE best practices.

    Known limitations

    Initial setup requires an environment with pipx or pip to run the full REUSE linting suite. GitHub Actions SHA resolution may require manual verification if network access is restricted.

    How to use

    Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scanned

    Verified clean 9 days ago

    • Free to download with an account

    Requires

    GitHub
    Python
    Node.js

    Listed9 days ago

    What's inside

    Frequently Asked Questions