self-improvement

Log learnings and errors to markdown files for continuous improvement.

1|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/mariusraileanu/openclaw-hardened-azure --skill self-improvement-mariusraileanu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/mariusraileanu/openclaw-hardened-azure/tree/main/workspace/skills/self-improving-agent
Command: npx skills add https://github.com/mariusraileanu/openclaw-hardened-azure --skill self-improvement-mariusraileanu

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Log learnings and errors to markdown files for continuous improvement. Coding agents can later process these into fixes, and important learnings get promoted to project memory.

Core Features & Use Cases

  • Log learnings and errors to .learnings/ files such as LEARNINGS.md, ERRORS.md, and FEATURE_REQUESTS.md
  • Promote important learnings to CLAUDE.md, AGENTS.md, and copilot-instructions.md
  • Support inter-session workflows and memory with memory logs and cross-session sharing
  • Enable periodic review and promotion of recurring patterns to workspace files

Quick Start

Create the .learnings directory and begin logging your first entry to the appropriate file.

Frequently Asked Questions about self-improvement

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

FAQPage Schema
How do I capture coding agent learnings and errors for cross-session memory?

To capture learnings and errors for cross-session memory, log entries to markdown files within a dedicated .learnings directory, utilizing LEARNINGS.md, ERRORS.md, and FEATURE_REQUESTS.md templates to structure the data.

How do I promote recurring learnings to project memory files like CLAUDE.md?

Promoting recurring learnings to project memory files like CLAUDE.md involves a frontmatter-based skill promotion workflow that elevates important entries from .learnings directory logs to CLAUDE.md, AGENTS.md, and copilot-instructions.md.

What is the best way to log continuous improvement workflows for long-running tasks?

The best way to log continuous improvement workflows for long-running tasks is capturing errors and corrections into structured markdown files, enabling coding agents to process these logs into fixes during periodic reviews.

Do I need a specific directory structure to start logging errors and feature requests?

Yes, you need a dedicated .learnings directory to start logging errors and feature requests, which serves as the central repository for the required LEARNINGS.md, ERRORS.md, and FEATURE_REQUESTS.md markdown templates.

Can I share knowledge and memory logs across multiple interactive prompt sessions?

Yes, you can share knowledge and memory logs across multiple interactive prompt sessions by capturing corrections and learnings to markdown files, enabling inter-session workflows and continuous improvement.

Why use markdown files for logging agent errors instead of relying on session context?

Logging agent errors to markdown files preserves knowledge beyond the immediate session context, allowing coding agents to process historical logs into fixes and promote recurring patterns to workspace memory.