A comprehensive template library providing five core design patterns to architect structured, reusable, and professional-grade AI agent skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-patterns
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-patterns 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 Patterns?
Skill Patterns is a specialized framework designed for developers looking to build high-quality Openclaw Skills. It serves as a blueprint for agent development, moving beyond simple prompting into structured software engineering for AI. By utilizing the Google ADK best practices, this skill enables agents to act as design experts, guiding users through the creation of robust skill architectures.
The library provides standardized templates and logical flows that ensure consistency across different agent capabilities. Whether you are building a simple utility or a complex multi-step automation, Skill Patterns provides the scaffolding necessary to manage domain expertise, output formatting, and quality control within the ecosystem of Openclaw Skills.
Skill Patterns Use Cases
- Creating a new AI agent skill from scratch using standardized templates.
- Optimizing existing Openclaw Skills to improve reliability and structure.
- Implementing automated code reviews or content auditing workflows.
- Designing complex, multi-step data transformation or generation pipelines.
- Gathering user requirements through structured interviews before executing tasks.
How Skill Patterns Works
- Requirement Analysis: The agent identifies user intent through trigger keywords like 'build a skill' or 'skill design'.
- Pattern Recommendation: Based on the goal, the agent recommends one or more of the 5 patterns (Tool Wrapper, Generator, Reviewer, Inversion, or Pipeline).
- Template Loading: The agent dynamically loads reference documents from the local directory to provide specific structural guidance.
- Structured Guidance: The user is guided through creating the necessary directory structure, including SKILL.md, references, and assets.
- Validation: The process concludes with a checklist to ensure the newly created skill meets technical and functional standards.
Skill Patterns Setup
To start building Openclaw Skills with these patterns, clone the repository into your agent's skill directory:
mkdir -p skills/skill-patterns
# Copy the skill files into the directory
# Ensure the references/ folder contains the pattern markdown files
Enable the skill by including it in your agent's configuration, ensuring the trigger keywords are active.
Skill Patterns Data Schema & Taxonomy
Skill Patterns follows a strict hierarchical directory structure to maintain modularity and reusability for all Openclaw Skills:
| Directory | File | Description |
|---|---|---|
/ |
SKILL.md |
The entry point defining triggers and core logic. |
references/ |
*.md |
Contains domain knowledge, checklists, and pattern definitions. |
assets/ |
template.md |
Reusable output formats and document scaffolding. |
scripts/ |
validate.py |
Optional scripts for automated validation or data processing. |
Skill Patterns Advanced Features
- Pattern Composition: Combine multiple patterns, such as using Inversion to gather data followed by a Pipeline for execution.
- Domain Expertise Injection: Use the Tool Wrapper pattern to force the agent to adhere to specific industry norms or team conventions.
- Self-Correction Loops: Integrate the Reviewer pattern with a Generator to create agents that automatically audit their own output before final delivery.
- Standardized ADK Alignment: Fully compatible with Google ADK Best Practices for professional agentic workflows.
SKILL.md
Loading
Related Openclaw Skills
HengshuiClaw
HengshuiClaw is a rigorous pressure-escalation framework designed to eliminate AI passivity and ensure agents exhaust every possible technical solution before conceding.
Li Lu Civilization Analysis
A strategic investment framework based on Li Lu's Civilization, Modernization, and Value Investing theory to evaluate long-term market trends.
Security Suite
A local-first reference tool providing instant access to security patterns, implementation guides, and performance best practices for developers.
SVG Optimizer Skill
A local, zero-dependency reference tool providing instant access to SVG optimization patterns, performance tips, and security best practices.
Wu Jun Investor Wisdom
A data-driven investment analysis tool based on the tech-forward principles of computer scientist and investor Wu Jun.
Terragrunt Infra Wrapper
A powerful utility toolkit for orchestrating, logging, and auditing Terragrunt and OpenTofu infrastructure operations.
Featured*
