Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A meta-skill that defines the authoritative standard for creating and updating OpenClaw skills. It enforces portable, reproducible skill structure with prerequisite checks, guided onboarding, and machine-specific config stored in skill-local files rather than baked into SKILL.md.
Enforces a consistent skill layout so any skill built to this standard works on another machine with minimal edits and no manual path-hunting.
When to use it
- Creating a new OpenClaw skill with standardized prerequisites and onboarding
- Auditing an existing skill for portability and reproducibility gaps
- Setting up a chat-first (Telegram-compatible) onboarding flow for a skill
- Splitting skill config into a shareable example file and a machine-specific real file
- Adding a smoke test step to validate skill setup after onboarding completes
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: create a new skill following skill-father standards
- 1Check prerequisites: verify required CLIs exist and any service accounts are accessible
- 2Create the skill folder with SKILL.md, config.env.example, and a scripts/ directory
- 3Write config.env.example with all required keys and safe defaults
- 4Implement onboarding: ask the user for machine-specific values in chat, write them to config.env
- 5Run a smoke test against the generated config and report pass or fail
A portable skill folder ready to share, with reproducible onboarding and no hardcoded machine-specific values