distill

Extract reusable lessons from conversational trial-and-error processes.

Updated Mar 4, 2026
One-click install
npx skills add https://github.com/xbghc/claude-plugins --skill distill-xbghc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: distill
Source: https://github.com/xbghc/claude-plugins/tree/main/plugins/ghm-skills/skills/distill
Command: npx skills add https://github.com/xbghc/claude-plugins --skill distill-xbghc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill captures valuable lessons learned through trial-and-error during a conversation, preventing the repetition of mistakes and codifying best practices.

Core Features & Use Cases

  • Identify Friction Points: Automatically scans conversation history for errors, retries, and workarounds.
  • Extract Reusable Knowledge: Distills these friction points into actionable insights.
  • Suggest Knowledge Codification: Recommends creating new skills, updating existing ones, or documenting findings in project files like CLAUDE.md.
  • Use Case: After a debugging session where you encountered and fixed several obscure errors, this skill can identify those errors and the solutions, then propose creating a new skill to document how to avoid them in the future.

Quick Start

Use the distill skill to review the current conversation for lessons learned.

Frequently Asked Questions about distill

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

FAQPage Schema
How do I extract reusable lessons learned from a debugging session?

You can extract reusable knowledge from debugging sessions by scanning conversation history for error-fix cycles and workarounds, then distilling these friction points into actionable insights and new skills.

What is the best way to document project-specific facts versus cross-project knowledge?

For project-specific facts, document findings in project files like CLAUDE.md; for cross-project applicability, prioritize creating or updating skills to capture reusable knowledge and prevent repeating mistakes.

Can I automatically identify friction points and knowledge gaps from my conversation history?

Yes, you can automatically identify friction points by scanning conversational history for error-fix cycles, solution switching, and knowledge gaps, then distilling these trial-and-error processes into actionable insights.

How do I codify insights from trial and error into new skills?

You codify insights by extracting reusable knowledge from friction points like error-fix cycles, then suggesting the creation of new skills for cross-project applicability or documenting project-specific facts in CLAUDE.md.

When should I update existing skills instead of creating new ones for knowledge management?

Update existing skills when extracted insights enhance already codified best practices for cross-project applicability, whereas creating new skills is prioritized for capturing distinct lessons learned from new friction points.