ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified developer tools Safety 4/5

marila-skill-publish

用于发布和更新 OpenClaw 技能到 ClawHub,并同步 GitHub Release。用户提到“发布技能”“发到 ClawHub”“发布这个 skill”“写完就发布”“上线这个技能”等场景时使用。包含完整发布步骤、版本规范、发布前检查清单、GitHub Release 同步规则和常见问题处理。

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/aliramw/marila-skill-publish
Or

What This Skill Does

The marila-skill-publish is an essential utility tool for OpenClaw developers designed to streamline the release lifecycle of AI agent skills. It automates the orchestration between your local development environment, the GitHub repository, and the ClawHub registry. This skill serves as a comprehensive workflow manager, ensuring that every version of your skill is properly documented, version-controlled via Git, and synced with official GitHub Releases. By enforcing strict adherence to metadata requirements and pre-publish checklists, it helps developers avoid common pitfalls such as missing environmental variable declarations or unstated binary dependencies, which are critical for the security and reliability of AI agents.

Installation

To begin using the publish workflow, ensure you have the required CLI tools installed on your development machine. The process requires Node.js (>=18), Git, and the GitHub CLI (gh). Install the core toolset using:

npm install -g clawhub
gh auth login

Once your environment is configured, initialize your skill project using standard Git practices and ensure your SKILL.md and package.json files correctly reflect the versioning and dependency requirements as outlined in the official ClawHub documentation. Always verify your identity with clawhub whoami before initiating the deployment process.

Use Cases

  • Production Releases: Use this workflow whenever you are pushing a new, stable version of your skill to the public registry.
  • Version Synchronization: When updating your skill, this tool ensures your CHANGELOG.md, package.json, and GitHub Release tags remain perfectly aligned.
  • Standardized CI/CD: Perfect for individual developers or teams looking to enforce a 'no release without a tag' policy, ensuring that users always have access to specific release notes and history.
  • Agent Workspace Deployment: Use it to safely sync your latest local changes to the ~/.openclaw/workspace/ directory for immediate testing.

Example Prompts

  1. "I've finished updating the documentation for my new web-scraper skill, please walk me through the steps to publish it to ClawHub."
  2. "I need to release version 1.2.0 of my utility skill; can you check my package.json and guide me through the GitHub Release and ClawHub sync process?"
  3. "I'm ready to ship this skill, let's run the pre-publish checklist to make sure I haven't missed any security metadata."

Tips & Limitations

  • Mandatory Checklist: Always review references/clawhub-review-checklist.md before executing the publish command. This is the most common point of failure for new developers.
  • Strict Syncing: Never skip the GitHub Release step. The system is designed to favor transparency; therefore, keeping your GitHub history consistent with your registry entries is a hard requirement.
  • Environment Safety: Be cautious when syncing files to ~/.openclaw/workspace/. Only perform this action in trusted environments to prevent accidental overwriting of sensitive configurations.
  • Documentation Integrity: Ensure your primaryEnv and requires.bins match exactly what is used in your code, as the ClawHub validator will reject inconsistent metadata.

Metadata

Author@aliramw
Stars4473
Views0
Updated2026-05-01
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-aliramw-marila-skill-publish": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#publishing#deployment#automation#workflow#developer-tools
Safety Score: 4/5

Flags: file-write, file-read, code-execution