Skill Authoring Guide for Openclaw

A technical framework for developing specialized AI agent capabilities through standardized documentation and modular script structures.

kjaylee
v1.0.0
Feb 9, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install skill-authoring

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install skill-authoring using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Skill Authoring Guide?

The Skill Authoring Guide is a specialized protocol designed to help developers extend the capabilities of AI coding agents. It provides a blueprint for creating Openclaw Skills that are both token-efficient and technically robust. By leveraging the agentskills.io standard, this guide ensures that agents receive domain-specific instructions and automated workflows without cluttering the context window with redundant information.

This skill focuses on the core principles of conciseness and freedom matching, allowing developers to define exactly how much autonomy an agent has during task execution. Whether you are building simple text-based instructions or complex multi-script automations, this authoring guide ensures your Openclaw Skills are optimized for high-performance agentic environments.

Skill Authoring Guide Use Cases

  • Creating new specialized workflows for AI coding agents that require domain-specific knowledge.
  • Optimizing existing Openclaw Skills to reduce token consumption while maintaining execution accuracy.
  • Standardizing organizational skill sets across a team to ensure consistent agent behavior.
  • Integrating local hardware or MiniPC tasks into an AI agent's available toolset via external delegation.

How Skill Authoring Guide Works

  1. Initialize a new directory within the local agent skills path and create a primary SKILL.md file.
  2. Define the skill identity using YAML frontmatter, ensuring the description is rich with trigger keywords that the agent can recognize.
  3. Determine the required level of autonomy—low freedom for strict scripts or high freedom for flexible text guidelines.
  4. Separate executable logic into a scripts folder and move detailed documentation to a references folder to keep the main instruction set lean.
  5. Deploy the skill into the Clawdbot environment where it is automatically indexed for use in active sessions.

Skill Authoring Guide Setup

To start authoring your own Openclaw Skills, set up the standard directory structure in your local environment:

# Navigate to your skills directory
cd ~/.clawdbot/skills

# Create a new skill structure
mkdir -p my-custom-skill/{scripts,references,assets}
touch my-custom-skill/SKILL.md

Ensure your SKILL.md file contains the mandatory YAML frontmatter with a unique name and descriptive metadata.

Skill Authoring Guide Data Schema & Taxonomy

The skill architecture follows a strict hierarchical structure to ensure portability and ease of indexing for Openclaw Skills:

Component Path Description
Core Logic /SKILL.md The primary entry point containing YAML metadata and instructions.
Executables /scripts/ Contains Python, Bash, or JS files for functional operations.
Documentation /references/ Detailed technical docs loaded only when specifically required.
Resources /assets/ Static files, templates, and images used by the skill.

Skill Authoring Guide Advanced Features

  • External task delegation: Ability to offload heavy computations or hardware tasks to a MiniPC using nodes.run commands.
  • Sub-agent spawning: Logic for skills to trigger and manage subordinate agents for complex multi-step workflows.
  • Dynamic context management: Guidelines for keeping SKILL.md under 500 lines to maximize token efficiency in large projects.
  • Secure script execution: Best practices for including trusted executable code within Openclaw Skills environments.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*