self-improvement

Log command failures, corrections, and feature requests to structured markdown files.

Updated Mar 2, 2026
One-click install
npx skills add https://github.com/baoxukang123/calculator-web --skill self-improvement-baoxukang123
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/baoxukang123/calculator-web/tree/main/skills/self-improving-agent
Command: npx skills add https://github.com/baoxukang123/calculator-web --skill self-improvement-baoxukang123

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps agents and developers capture learnings, errors, and feature requests to enable continuous improvement and knowledge sharing within a project.

Core Features & Use Cases

  • Error Logging: Records command failures and unexpected behaviors.
  • Learning Capture: Logs corrections, insights, and knowledge gaps.
  • Feature Request Tracking: Documents missing capabilities.
  • Use Case: When a script fails with an obscure error, log it to .learnings/ERRORS.md. If the user corrects the agent's output, log it as a learning. If a recurring pattern emerges, promote it to project memory files like CLAUDE.md or AGENTS.md.

Quick Start

Log a command failure by running the script with the error details.

Frequently Asked Questions about self-improvement

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

FAQPage Schema
How do I track agent errors and log command failures for debugging?

You can log command failures by recording error details and unexpected behaviors in structured markdown files. This captures detailed context and metadata for continuous agent debugging and improvement.

What is the best way to capture knowledge gaps and user corrections in a project?

The best way to capture knowledge gaps and user corrections is to log them as structured learnings in markdown files. This records insights and corrections for future reference and agent improvement.

How do I save feature requests and missing capabilities during development?

You can save feature requests and missing capabilities by documenting them in structured markdown files. This tracks missing capabilities with detailed context and metadata for future implementation.

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

Yes, you can promote recurring learnings to project memory files like CLAUDE.md or AGENTS.md. When a recurring pattern emerges, promoting it enables broader application across the project.

How do I structure error logs and learnings for continuous improvement?

You structure error logs and learnings by saving them as structured markdown files with detailed context and metadata. This format supports logging command failures, corrections, and knowledge gaps.