compound-docs

Capture verified technical solutions into categorized markdown docs with validated YAML frontmatter.

9|Updated Mar 9, 2026
One-click install
npx skills add https://github.com/sylvanding/omelette --skill compound-docs-sylvanding
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: compound-docs
Source: https://github.com/sylvanding/omelette/tree/main/.claude/skills/compound-docs
Command: npx skills add https://github.com/sylvanding/omelette --skill compound-docs-sylvanding

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill eliminates the loss of hard-earned problem-solving knowledge when team members change roles or sessions end, by automatically capturing verified technical solutions into a structured, searchable knowledge base.

Core Features & Use Cases

  • Automatic Solution Capture: Triggers automatically when confirmation phrases like "that worked" are detected, or via the manual /doc-fix command, to document non-trivial solved problems.
  • Validated Structured Documentation: Uses enum-validated YAML frontmatter and standardized templates to ensure all solution docs are consistent, searchable, and categorized correctly.
  • Cross-Referencing & Pattern Detection: Links related issues and identifies recurring critical patterns to promote to required reading for the whole team.
  • Use Case: When a developer fixes a hard-to-debug N+1 query issue in a Rails application, the skill automatically creates a categorized doc with exact error messages, root cause, solution code, and prevention guidance that any team member can find later.

Quick Start

After confirming a problem is solved, say "that worked" to automatically trigger the skill to document the solution, or run the /doc-fix command to start the process manually.

Frequently Asked Questions about compound-docs

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

FAQPage Schema
How do I automatically document solved software bugs for my team knowledge base?

You can document solved software bugs by triggering capture with confirmation phrases like "that worked" or the manual /doc-fix command. This automatically generates structured markdown documentation with validated YAML frontmatter to retain troubleshooting knowledge.

How do I structure troubleshooting documentation to make it searchable across sessions?

Structure troubleshooting documentation using standardized templates with enum-validated YAML frontmatter. This ensures solution docs are consistently formatted, correctly categorized, and fully searchable across different team members and development sessions.

What's the best way to retain institutional knowledge of Rails bug fixes when developers leave?

Retain institutional knowledge of Rails bug fixes by capturing verified solutions into a categorized knowledge base. The system records exact error messages, root causes, solution code, and prevention guidance that any remaining team member can easily find later.

Can I link related debugging issues and identify recurring critical patterns in my workflow?

Yes, you can link related debugging issues and identify recurring critical patterns. The system cross-references related issues and promotes detected critical patterns to required reading, ensuring the entire team is aware of frequent problems.

Does this knowledge base tool require any specific dependencies or environments to run?

No, this knowledge base tool requires no external dependencies to run. It operates independently to capture verified technical solutions into structured markdown files, relying only on its internal assets and references components.

When should I not use automated documentation capture for my software development workflow?

Avoid using automated documentation capture for trivial problems that lack debugging value. The system is designed to trigger specifically for non-trivial solved problems to ensure your knowledge base retains high-quality, meaningful troubleshooting data.