Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Step 11 now prepares publish package and logs D-## approval request — does NOT execute publish. Step 11.5 is the actual publish execution, gated on explicit D-## yes. ClawhHub publish is now a hard-gated external action.
What This Skill Does
Manages the full publication pipeline for releasing skills to ClawhHub. Handles structure scaffolding, OPSEC scanning, staging, user review, and public release. Batch releases run Phase 1 in parallel and collapse to a single review message; git history is erased before the repo goes public.
Collapses scaffolding, OPSEC scanning, staging, review, history erasure, and scan verification into a single command with one human approval gate.
When to use it
- Publishing a finished skill to ClawhHub from a local repo
- Scanning a skill directory for sensitive data before public release
- Batch-releasing multiple skills with one approval message
- Checking whether a skill meets all readiness requirements
- Reviewing a staged skill on a private GitHub repo before flipping it public
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Release skill-engineer to ClawhHub
- 1Scaffolds missing files (README.md, CHANGELOG.md, LICENSE, tests/) from SKILL.md
- 2Runs OPSEC scan on release copy and creates a private GitHub staging repo
- 3Pushes sanitized content and sends user a review link with the private repo URL
- 4On approval, erases git history via orphan branch and flips repo to public
- 5Publishes to ClawhHub, verifies metadata matches skill.yml, checks VirusTotal scan result
Delivery message with GitHub URL, ClawhHub listing URL, and security scan verdict
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.