Improvement Learner provides a structural quality assessment and self-improvement loop for Openclaw Skills using a 6-dimensional evaluation framework.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install auto-improvement-learner
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 auto-improvement-learner 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 Improvement Learner?
Improvement Learner is a technical tool built to implement a self-improvement loop for Openclaw Skills, following the evaluate-modify-re-evaluate cycle. It focuses on the structural integrity and quality of skill definitions, ensuring that every SKILL.md file meets high standards for accuracy and efficiency. By using a Pareto front approach, the tool guarantees that no dimension of the skill regresses during the optimization process.
This tool is critical for developers who want to maintain professional-grade Openclaw Skills. It moves beyond simple text editing by utilizing a three-layer memory system (HOT, WARM, and COLD) to store and retrieve optimization patterns, ensuring that the AI agent learns from previous iterations to produce better results over time.
Improvement Learner Use Cases
- Evaluating a skill across six distinct quality dimensions to find structural weaknesses.
- Running automated self-improvement cycles to optimize Openclaw Skills documentation and triggers.
- Monitoring the historical score variations of a skill to ensure long-term quality maintenance.
- Detecting security risks such as hardcoded API keys or unsafe function calls within skill definitions.
How Improvement Learner Works
- The tool performs an initial evaluation of the SKILL.md file against 15 specific accuracy items and trigger quality metrics.
- It calculates coverage by checking for the presence of scripts, tests, and README files that support the core skill.
- Reliability is tested via pytest execution, while efficiency is measured based on optimal line count thresholds.
- A security scan is conducted to identify sensitive credentials or dangerous code execution patterns.
- The tool enters a modification loop where it proposes changes to improve scores.
- Improvements are only committed if they satisfy the Pareto front criteria, ensuring no single dimension is sacrificed for another.
Improvement Learner Setup
To integrate Improvement Learner into your workflow for Openclaw Skills, use the included Python scripts. Ensure you have the necessary environment variables and paths configured.
# Evaluate a specific skill to view its current quality scores
python3 scripts/self_improve.py --skill-path /path/to/skill --max-iterations 1
# Run a full self-improvement cycle with memory tracking
python3 scripts/self_improve.py \
--skill-path /path/to/skill \
--max-iterations 5 \
--memory-dir /path/to/memory \
--state-root /path/to/state
# Export historical progress data to a JSON file
python3 scripts/track_progress.py --skill-path /path/to/skill --output progress.json
Improvement Learner Data Schema & Taxonomy
Improvement Learner generates structured JSON artifacts to document the evolution of Openclaw Skills quality. The data is organized as follows:
| Artifact | Contents |
|---|---|
| Evaluation JSON | 6-dimension scores ranging from 0.0 to 1.0 for each category. |
| Improvement Log | Details on iterations, specific changes kept or reverted, and memory stats. |
| Progress Report | Historical trend data showing how scores have changed over multiple versions. |
| Memory State | Pattern data stored in HOT (frequent), WARM (on-demand), or COLD (archive) layers. |
Improvement Learner Advanced Features
- Pareto Front Protection: Ensures that automated updates to Openclaw Skills never degrade existing quality scores.
- Three-Layer Memory Architecture: Optimizes performance by categorizing learning patterns based on access frequency.
- Structural Accuracy Checks: Validates 15 specific points including frontmatter, code examples, and atomicity.
- Trend Analysis: Provides deep insights into the lifecycle of a skill through automated historical tracking.
SKILL.md
Loading
Related Openclaw Skills
Improvement Executor
A precision execution engine that applies approved AI candidates to target files with built-in safety guards and automatic backup systems.
ByteDance Organization Framework
A management methodology skill based on the ByteDance philosophy of Context not Control and algorithm-driven organizational systems.
Feelings Framework
A persistent emotional layer for AI agents that tracks mood and feeling intensities over time to shape dynamic response behaviors.
Improvement Gate
A multi-layered validation framework that protects code integrity by enforcing strict mechanical quality gates on AI-generated improvements.
Improvement Generator
An automated engine that produces ranked improvement candidates for AI skills based on feedback loops, failure traces, and target analysis.
Lei Jun Strategic Mindset
A strategic AI skill that applies Lei Jun's five core mental models to evaluate products, business trends, and operational efficiency.
Featured*
