self-improvement

Log learnings, errors, and corrections to .learnings markdown files.

Updated Apr 4, 2026
One-click install
npx skills add https://github.com/SimonOneNineEight/openclaw --skill self-improvement-simononenineeight
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/SimonOneNineEight/openclaw/tree/main/workspace/skills/self-improving-agent
Command: npx skills add https://github.com/SimonOneNineEight/openclaw --skill self-improvement-simononenineeight

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

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User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task.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.

Core Features & Use Cases

  • Log learnings to .learnings/LEARNINGS.md
  • Log errors to .learnings/ERRORS.md
  • Log feature requests to .learnings/FEATURE_REQUESTS.md
  • Promote broadly applicable learnings to CLAUDE.md, AGENTS.md, or .github/copilot-instructions.md
  • Integrates with OpenClaw workspace for memory and agent workflows

Quick Start

Log to .learnings/ after a failure, correction, or new capability request.

Frequently Asked Questions about self-improvement

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I log errors and corrections during OpenClaw sessions?

You can log errors and corrections by writing entries to .learnings/ERRORS.md or .learnings/LEARNINGS.md using structured frontmatter and markdown templates to enable continuous improvement.

What is the best way to capture learnings from command failures and outdated knowledge?

The best way to capture learnings from command failures and outdated knowledge is to log them into structured markdown templates with frontmatter, storing entries in a dedicated .learnings folder.

How do I promote recurring patterns to CLAUDE.md or AGENTS.md?

You promote recurring patterns to CLAUDE.md or AGENTS.md by elevating broadly applicable learnings from the local .learnings folder into these central instruction files.

Can I log feature requests and user corrections in the .learnings folder?

Yes, you can log feature requests and user corrections by adding entries to .learnings/FEATURE_REQUESTS.md and .learnings/LEARNINGS.md using structured frontmatter and markdown templates.

When should I review learnings before major tasks in OpenClaw?

You should review learnings before major tasks in OpenClaw to prevent repeating past errors, applying logged corrections and recurring patterns stored in the .learnings folder.