A comprehensive AI programming orchestrator that manages the full task lifecycle, from code generation to automated knowledge induction.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install evolving-agent
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 evolving-agent 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 Evolving Agent?
Evolving Agent serves as a master process supervisor designed to coordinate complex AI-driven development workflows. By leveraging these Openclaw Skills, the agent acts as a central nervous system that identifies user intent—whether it be active programming, knowledge organization, or learning from external sources—and delegates tasks to specialized sub-modules.
Built on the principle of progressive disclosure, the skill maintains a clean orchestration layer while offloading detailed implementation to modules like the programming-assistant and knowledge-base. It ensures that every action is part of a verified, closed-loop system that not only completes the task but also captures the experience for future use.
Evolving Agent Use Cases
- Automating the end-to-end lifecycle of software features, including creation, bug fixing, and refactoring.
- Building a persistent project knowledge base by automatically extracting patterns and lessons learned during development.
- Learning new coding styles and architectural patterns by analyzing and fetching data from GitHub repositories.
- Implementing multi-step agentic workflows that require constant health monitoring and status verification.
How Evolving Agent Works
- Set the environment path variables to locate the relevant Openclaw Skills modules on the local system.
- Utilize sequential-thinking tools to perform deep intent recognition, classifying requests into Programming, Induction, or Learning categories.
- Automatically load the corresponding module documentation and scripts based on the identified intent.
- Execute sub-processes for the specific task while maintaining a master supervisor role to prevent logic drift.
- Monitor execution health by reading internal project trackers such as progress logs and task status JSON files.
- Conduct a final audit of all outputs to ensure task completion, code quality, and successful experience extraction.
Evolving Agent Setup
To integrate Evolving Agent into your workflow, ensure your Openclaw Skills directory is correctly configured and the runner script is accessible.
# Define the skills directory path
SKILLS_DIR=$([ -d ~/.config/opencode/skills/evolving-agent ] && echo ~/.config/opencode/skills || echo ~/.claude/skills)
# Initialize the evolution mode to start tracking progress
python $SKILLS_DIR/evolving-agent/scripts/run.py mode --init
# Verify the current status of the agent
python $SKILLS_DIR/evolving-agent/scripts/run.py mode --status
Evolving Agent Data Schema & Taxonomy
The skill manages its state and metadata using a structured taxonomy within the .opencode directory. This allows for persistent tracking across sessions.
| File | Type | Description |
|---|---|---|
progress.txt |
Text | Real-time log of the current execution steps. |
feature_list.json |
JSON | A structured list of tasks and their current states (pending, completed, blocked). |
.evolution_mode_active |
Flag | Marker file indicating that the agent is in automatic experience extraction mode. |
Evolving Agent Advanced Features
- Intent-based module delegation for Programming, Knowledge Induction, and GitHub Learning.
- Automated health checks that monitor task dependencies and blockages in real-time.
- Closed-loop verification system that requires all tasks to reach a completed status before finalization.
- Evolution Mode which triggers automatic extraction of development patterns into the local knowledge base.
- Seamless integration with the sequential-thinking tool for complex logic decomposition.
SKILL.md
Loading
Related Openclaw Skills
Execution Verifier
A skill that prevents fake progress by enforcing artifact-based verification and hard evidence reporting for AI agent tasks.
News Writing
A comprehensive journalistic workflow for AI agents to collect materials, verify facts, and generate publication-ready news content.
Google Analytics
A comprehensive integration for managing Google Analytics 4 properties and retrieving reporting data via managed OAuth.
AI Voice Agent Backend
A local-to-cloud bridge for processing Speech-to-Text with Whisper and Text-to-Speech via AWS Polly.
Ralph Mode: Autonomous Development Loops
An autonomous development framework for Openclaw Skills that uses iterative loops, validation gates, and structured planning to manage complex coding tasks.
QWeather City Weather
A portable Python-based skill for resolving city IDs and fetching real-time weather data via the QWeather API.
Featured*
