A powerful framework for the automated generation, templating, and composition of dynamic AI agents and system prompts.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-generator
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 agent-generator 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 Agent Generator?
The Agent Generator is a specialized skill designed for developers who need to architect sophisticated AI systems through automated agent creation and advanced metaprogramming. By leveraging a robust technical stack including template engines like Handlebars and DSL design tools, this skill allows for the systematic construction of agents based on modular capabilities and patterns.
Integrating this tool with Openclaw Skills enables the transition from static, hand-written prompts to dynamic, self-improving architectures. It provides the necessary infrastructure to handle complex requirements such as capability composition, pattern recognition, and automated optimization, ensuring that every generated agent is both performant and maintainable within a larger ecosystem.
Agent Generator Use Cases
- Scaffolding domain-specific agents using modular templates and standardized patterns.
- Creating self-modifying or adaptive AI systems that generate new agent instances based on runtime logic.
- Standardizing prompt engineering across large teams using centralized template libraries.
- Composing complex multi-agent workflows by inheriting and combining distinct capability sets.
How Agent Generator Works
- Analyze the specific domain requirements to identify the necessary agent patterns and capability mappings.
- Define the agent structure using a declarative Domain-Specific Language (DSL) or JSON Schema.
- Utilize the integrated template engines to inject logic into system prompts and code scaffolds.
- Apply capability composition to build complex behaviors from simpler, reusable components within the Openclaw Skills environment.
- Validate the generated output using static analysis, type inference, and AST manipulation to ensure code quality.
- Deploy the generated agent and iterate based on performance benchmarks and capability analysis.
Agent Generator Setup
To begin using the Agent Generator within your project, ensure your environment is configured for TypeScript and the relevant template engines. Use the following commands to install the core generation dependencies:
# Install essential generation and templating tools
npm install typescript handlebars ejs babel-core
# Initialize the agent template directory
mkdir -p templates/agents capability-maps/
Agent Generator Data Schema & Taxonomy
The skill organizes its generation logic through a structured schema to ensure consistency across all Openclaw Skills implementations:
| Component | File Type | Purpose |
|---|---|---|
| Agent Templates | .hbs, .j2, .ejs | Blueprints for system prompts and agent logic |
| Capability Map | .json | Definitions of modular skills and tools |
| DSL Definitions | .ts, .pegjs | Grammar and syntax for declarative agent design |
| Validation Rules | .schema.json | JSON Schema for structural verification |
Agent Generator Advanced Features
- Advanced capability inheritance for building hierarchical agent structures.
- Support for multiple template engines including Liquid, Mustache, and Jinja2.
- Automated prompt optimization based on property-based and mutation testing.
- Seamless integration with LangChain and CrewAI for multi-agent orchestration.
- Dynamic code generation via TypeScript Compiler API and AST manipulation.
- Built-in performance benchmarking for generated Openclaw Skills to ensure efficiency.
SKILL.md
Loading
Related Openclaw Skills
Accessibility Auditor
The Accessibility Auditor is a specialized AI persona focused on achieving WCAG 2.1/2.2 compliance through automated testing, manual procedures, and inclusive design practices.
Humanizer-Pro
Humanizer-Pro is a high-end AI writing purification engine that goes beyond simple de-noising to reconstruct text with cognitive depth and emotional resonance.
Tibber Energy
An AI-powered energy management skill that leverages Tibber API data to automate smart home actions and optimize electricity consumption based on real-time spot prices.
Solar-Tilt-Calculator
A professional-grade solar radiation calculator that transforms NASA climatology data into actionable tilted surface radiation metrics.
Agent Communicator
A specialized communication layer that enables AI agents to coordinate work through standardized message-passing protocols and structured handoffs.
AD Security Reviewer
An AI-powered Active Directory security specialist that identifies identity attack paths and provides actionable hardening recommendations for enterprise domains.
Featured*
