self-improvement

Analyze local Claude conversation logs to surface recurring friction patterns.

10|3|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/bokan/claude-skill-self-improvement --skill self-improvement-bokan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-improvement
Source: https://github.com/bokan/claude-skill-self-improvement/tree/main
Command: npx skills add https://github.com/bokan/claude-skill-self-improvement --skill self-improvement-bokan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill analyzes local Claude conversation histories to surface recurring friction patterns, user frustration points, and gaps in CLAUDE.md or skill configuration so maintainers can prioritize targeted improvements.

Core Features & Use Cases

  • Parallel analysis of conversation .jsonl logs to extract user goals, failures, what worked, and representative user quotes for evidence-based insight.
  • Synthesis into ranked friction patterns, suggested CLAUDE.md additions, a registry of already-fixed items, and candidate new skills to build.
  • Use Case: Run against a project's recent sessions to reveal repeated misunderstandings, propose exact documentation changes, and shortlist skills that would prevent repeat issues.

Quick Start

Run the self-improvement skill to analyze recent conversation logs and generate CLAUDE_IMPROVEMENTS.md for review.

Frequently Asked Questions about self-improvement

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

FAQPage Schema
How do I analyze conversation history logs to find recurring friction patterns?

Analyzing conversation history logs for friction patterns involves parsing local Claude .jsonl archives to extract user goals, failures, and successful prompts. This surfaces recurring frustrations and generates a ranked summary of friction points for targeted project improvements.

How do I generate CLAUDE.md config suggestions from previous chat sessions?

Generating CLAUDE.md config suggestions from chat sessions requires analyzing local conversation .jsonl files to identify documentation gaps. The analysis produces exact proposed additions to your configuration file, addressing repeated misunderstandings and missing instructions.

What is the best way to identify missing skills from Claude conversation logs?

Identifying missing skills from Claude conversation logs is best achieved by synthesizing friction patterns from .jsonl session archives. This analysis shortlists potential new skills that would prevent repeat issues and user frustrations.

Does friction detection work on local files without making external network requests?

Yes, friction detection operates entirely on local Claude project log files and conversation .jsonl archives without making external network requests. This ensures your session data remains private while extracting failures and successful prompts.

How do I extract successful prompts and fixed items from .jsonl conversation archives?

Extracting successful prompts and fixed items from .jsonl conversation archives requires parallel analysis of local session logs. This process builds a registry of already-fixed items and identifies representative user quotes for evidence-based insight.