A specialized AI creation coach that guides developers and experts through the process of packaging professional capabilities into reusable Openclaw Skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-coach
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install skill-coach 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 Coach?
Skill Coach is an expert-level framework designed to bridge the gap between human expertise and executable AI capabilities. Rather than simply generating code, it acts as a mentor that teaches users how to design robust Openclaw Skills through a structured, conversational interface. It emphasizes clear constraints, flexible implementation, and concise documentation to prevent bloated or inefficient AI workflows.
By utilizing diagnostic discovery and adaptive complexity modes, Skill Coach ensures that every created skill is purpose-built for its specific use case. It adheres to the Minimal Viable Principle, focusing on making the skill functional first before layering in advanced domain knowledge or technical integrations, ensuring a high success rate for both simple and complex automation tasks.
Skill Coach Use Cases
- Converting repetitive manual workflows into automated Openclaw Skills for AI agents.
- Standardizing domain-specific knowledge into structured formats for team-wide AI usage.
- Designing multi-step decision engines for regulated industries like finance and healthcare.
- Rapidly prototyping new AI capabilities using Simple, Standard, or Deep development modes.
How Skill Coach Works
- Discovery & Scenario Parsing: The coach identifies the recurring task and evaluates its suitability for automation based on frequency and value.
- Complexity Assessment: The system automatically infers whether the skill requires a Simple (3-step), Standard (4-step), or Deep (5-step) development flow.
- Boundary Definition: Users define explicit inputs, outputs, and triggers to establish clear operational constraints for the AI agent.
- Structured Development: The coach drafts the SKILL.md file and organizes supporting assets like scripts, references, or MCP tool definitions.
- Verification & Delivery: The final files are generated with a built-in validation checklist to ensure the new Openclaw Skills are ready for immediate deployment.
Skill Coach Setup
To initialize the Skill Coach within your environment, ensure you have the Openclaw Skills directory structure ready. You can interact with the coach to begin the creation process:
# Navigate to your skills directory
cd my-ai-project/skills/
# Create a new skill directory using kebab-case
mkdir skill-coach-session
# Query existing cases from the knowledge base to get started
python3 knowledge-base/scripts/query_cases.py --list
Skill Coach Data Schema & Taxonomy
The skill follows a strict hierarchical file structure to ensure modularity and ease of maintenance:
| Directory | Purpose | Mandatory |
|---|---|---|
SKILL.md |
The core definition and logic of the skill. | Yes |
scripts/ |
Python executors for data processing or API calls. | No |
references/ |
Markdown files containing domain-specific guidelines. | No |
assets/ |
Templates (Word, Excel, Email) and static resources. | No |
tools/ |
JSON definitions for Model Context Protocol (MCP) tools. | No |
All metadata is stored in the SKILL.md header using YAML front matter, including the skill name, description, and specific trigger keywords.
Skill Coach Advanced Features
- Adaptive Complexity Logic: Automatically scales the development workflow based on the number of inputs and decision branches.
- Knowledge Base Integration: Accesses a SQLite-backed repository of best practices and edge cases via specialized Python scripts.
- Industry Compliance Guardrails: Automatically injects mandatory disclaimers and data timeliness warnings for healthcare, legal, and financial sectors.
- Step Confirmation Mechanism: Implements a 'Pause and Confirm' protocol at critical nodes to ensure the generated logic aligns perfectly with user intent.
- Multi-Agent Support: Allows splitting complex decision trees (depth > 5) into sub-skills for better maintainability.
SKILL.md
Loading
Related Openclaw Skills
Handyman Video
A specialized AI agent skill designed to transform handyman reliability and multi-trade expertise into high-converting trust videos.
address2lnglat
A Python-powered utility for batch converting Chinese location names into precise BD-09 Baidu coordinates using official APIs.
Performance Analysis & Optimization
A powerful performance analysis tool to profile code execution, identify architectural bottlenecks, and provide actionable optimization suggestions.
Code Review Assistant
A specialized tool for automating code quality analysis, security vulnerability detection, and adherence to best practices across multi-language projects.
Duhui Document to PDF Converter
A high-performance async conversion tool that transforms local Office, WPS, and image files into professional PDF documents.
Where Not Crowded Report
A travel decision-making skill that analyzes crowd trends and costs to suggest peaceful, high-value alternatives to popular holiday destinations.
Featured*
