Skill Evolver for Openclaw

A comprehensive skill lifecycle manager designed to discover, orchestrate, and evolve reusable AI agent capabilities.

testlbin
v0.1.0
Mar 10, 2026
1
1.6k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install skill-evolver

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-evolver 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 Evolver?

Skill Evolver is a sophisticated framework for managing the entire lifecycle of Openclaw Skills within an AI agent environment. It serves as a strategic layer that helps developers decide when to use existing tools, how to compose multiple capabilities into a single workflow, and when to materialize a successful execution path into a brand new, reusable skill.

By following a structured methodology of solve first, materialize later, this tool ensures that your AI agent remains lean and efficient. It prioritizes native abilities and simple orchestration over unnecessary complexity, only creating persistent Openclaw Skills when there is proven reuse value and validated success in real-world scenarios.

Skill Evolver Use Cases

  • Orchestrating multiple Openclaw Skills to complete a complex multi-step task that no single tool can handle.
  • Discovering and auditing new capabilities from a local or remote registry.
  • Fusing partially matching skills into a new, specialized capability to fill functional gaps.
  • Preserving a successful manual workflow by converting it into a permanent, reusable skill for future sessions.
  • Conducting security audits and deep inspections of third-party tools before integration.

How Skill Evolver Works

  1. Setup and Intent Analysis: The system initializes a timestamped directory and analyzes the user's request to define the specific features needed.
  2. Skill Discovery: It performs a dual-track search across local environments and the Openclaw Skills registry to identify candidate tools.
  3. Deep Inspection: Each candidate is audited for security and capability alignment to determine if it meets the technical requirements.
  4. Approach Decision: The LLM recommends a path: simple orchestration, skill fusion for custom needs, or falling back to native abilities.
  5. Execution and Orchestration: A formal execution plan is generated and confirmed by the user before the agent invokes the necessary tools.
  6. Materialization: After a successful outcome, the user is prompted to decide if the workflow should be saved as a new, persistent skill.

Skill Evolver Setup

To begin managing your Openclaw Skills, initialize your workspace and set your output path:

# Create a session-specific directory
mkdir -p "output/$(date +%m-%d)-your-feature-slug"

# Set the environment variable for the session
OUTPUT_DIR=output/$(date +%m-%d)-your-feature-slug

Ensure you have access to the skill-search and skill-creator components within your AI protocol environment.

Skill Evolver Data Schema & Taxonomy

The Skill Evolver organizes session data using a structured directory hierarchy and specific markdown templates to track the evolution of Openclaw Skills:

File Path Description
01-intent.md Initial task analysis and requirement mapping.
02-candidates.md Merged search results from local and registry tracks.
02-audit.md Security and permission audit reports for selected tools.
03-inspection.md Detailed technical analysis of skill manifests and schemas.
04-orchestration.md The final execution plan and sequence of tool invocations.

Skill Evolver Advanced Features

  • Dual-Track Search: Simultaneously queries local repositories and the global registry for Openclaw Skills.
  • Human-in-the-Loop Checkpoints: Integrated 'AskUserQuestion' prompts at critical decision points to ensure alignment before execution.
  • Conditional Fusion: Sophisticated logic to invoke a skill-creator when existing tools only partially satisfy a requirement.
  • Materialization Engine: Automatically transforms validated session logs into standardized skill definitions for long-term use.
  • Security Audit Layer: Mandatory security reporting phase for every new tool discovered during the search process.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*