self-improvement

Log learnings and errors into .learnings markdown files for continuous improvement.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/rumihut/job-researcher-landing --skill self-improvement-rumihut
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/rumihut/job-researcher-landing/tree/main/skills/self-improving-agent
Command: npx skills add https://github.com/rumihut/job-researcher-landing --skill self-improvement-rumihut

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 across development sessions, ensuring insights are preserved and reusable.

Core Features & Use Cases

  • Central logging for corrections, errors, knowledge gaps, and best practices.
  • Automatic log file setup in the project root (.learnings/LEARNINGS.md, .learnings/ERRORS.md, .learnings/FEATURE_REQUESTS.md) and structured entries.
  • Hook-driven reminders to capture learnings after prompts or tool runs, plus promotion pathways to CLAUDE.md, AGENTS.md, and copilot-instructions.md for broadly applicable learnings.

Quick Start

Create the .learnings directory with LEARNINGS.md, ERRORS.md, and FEATURE_REQUESTS.md if missing, and log learnings and errors as they happen.

Frequently Asked Questions about self-improvement

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

FAQPage Schema
How do I capture learnings and errors for continuous improvement across development sessions?

To capture learnings and errors for continuous improvement, initialize a `.learnings` directory with `LEARNINGS.md`, `ERRORS.md`, and `FEATURE_REQUESTS.md`, then log standardized entries when commands fail or corrections are made.

When does the self-improvement logging mechanism trigger an entry?

The logging mechanism triggers entries when commands fail, users correct outputs, external tools fail, knowledge is outdated, missing features are requested, or better approaches are discovered during development.

How do I promote logged learnings to my AI coding assistant instructions?

To promote logged learnings to assistant instructions, move standardized entries from the `.learnings` log files into `CLAUDE.md`, `AGENTS.md`, or `copilot-instructions.md` so broadly applicable insights are preserved and reusable.

Do I need to manually set up the logging directory before capturing errors?

Yes, you need to manually create the `.learnings` directory containing `LEARNINGS.md`, `ERRORS.md`, and `FEATURE_REQUESTS.md` in your project root before the hook-driven reminders can log structured entries.

Can I use hook-driven reminders to automatically log corrections in my project?

Yes, you can use hook-driven reminders to automatically prompt logging after tool runs or user prompts, ensuring corrections and errors are captured in the standardized `.learnings` directory format.

What is the best way to track missing features and knowledge gaps during development?

The best way to track missing features and knowledge gaps is logging them into `FEATURE_REQUESTS.md` and `LEARNINGS.md` within a `.learnings` directory, using hook-driven reminders to capture entries as they occur.