chen-self-improvement
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
Why use this skill?
Enhance your OpenClaw agent with chen-self-improvement. Capture errors, corrections, and best practices into structured logs for a smarter, adaptive AI workflow.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/cs995279497-byte/chen-self-improvementWhat This Skill Does
The chen-self-improvement skill is a sophisticated meta-cognition tool for OpenClaw agents, designed to transform fleeting interactions into a persistent, evolving knowledge base. By capturing errors, user corrections, and successful optimizations in structured markdown logs, the skill enables your AI agent to learn from its past, effectively reducing repetitive mistakes and enhancing operational efficiency over time. It acts as the agent's institutional memory, ensuring that once a lesson is learned, it is codified in the local workspace, allowing the agent to reference, iterate, and improve upon its prior performance during future tasks.
Installation
To integrate this skill, use the recommended OpenClaw installer via ClawdHub: clawdhub install self-improving-agent. Alternatively, for manual control, you can clone the repository from https://github.com/peterskoett/self-improving-agent.git into your ~/.openclaw/skills/ directory. Once installed, ensure your workspace includes the .learnings/ directory containing LEARNINGS.md, ERRORS.md, and FEATURE_REQUESTS.md files to allow the agent to read and write its experiences.
Use Cases
This skill is indispensable for long-term development tasks. Use it when an API integration fails to provide necessary context, allowing the agent to log the specific error and prevent future retries with incorrect payloads. It is ideal for iterative coding sessions where you provide feedback on stylistic preferences or logic structure, ensuring your preferences are remembered. It also excels in discovering better workflows; when you find a faster way to accomplish a goal, logging it as a 'best_practice' allows the agent to adopt that sequence as a default behavior.
Example Prompts
- "I corrected you on the API endpoint format. Please log this to the learnings file as a knowledge_gap so you don't make the same mistake next time."
- "We found that using the --force flag is risky for this database schema. Add this as a best practice in our learnings logs."
- "That last command failed with a timeout. Can you document the error in the ERRORS.md file so we can analyze the pattern of these connection drops?"
Tips & Limitations
To maximize the utility of this skill, treat your workspace files as a living document. Regularly review LEARNINGS.md to identify recurring issues that warrant promotion to AGENTS.md or SOUL.md for permanent behavioral adjustments. A key limitation is that the skill relies on the agent's ability to interpret and trigger writes; ensure the agent has write permissions for the ~/.openclaw/workspace/.learnings/ directory. Do not store sensitive credentials in these logs, as they are saved in plain text files within your local environment.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-cs995279497-byte-chen-self-improvement": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read