compound-docs

Convert solved research problems into Markdown files with YAML frontmatter.

11|2|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/James-Traina/compound-science --skill compound-docs-james-traina
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: compound-docs
Source: https://github.com/James-Traina/compound-science/tree/main/skills/compound-docs
Command: npx skills add https://github.com/James-Traina/compound-science --skill compound-docs-james-traina

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill captures solved research problems and converts them into structured, searchable documentation using YAML frontmatter, enabling rapid reuse in future sessions.

Core Features & Use Cases

  • Categorized documentation: each solved problem is saved as a single Markdown file under a category directory (e.g., docs/solutions/estimation-issues/...).
  • Frontmatter-driven searchability: metadata fields like component, date, category, symptoms, root_cause, and solution enable precise querying and pattern detection.
  • Future-proof knowledge base: documents can be used by agents like learnings-researcher to accelerate subsequent analyses and reproduce results.

Quick Start

Document a solved research problem by creating a categorized Markdown file under docs/solutions with YAML frontmatter.

Frequently Asked Questions about compound-docs

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

FAQPage Schema
How do I document solved research problems for future lookup?

Structured documentation captures solved research problems as categorized Markdown files with YAML frontmatter, using metadata fields like component, date, category, symptoms, root_cause, solution, and prevention to enable precise querying and pattern detection for future lookup.

What YAML frontmatter fields are needed for a research knowledge base?

Building a research knowledge base requires YAML frontmatter fields including component, date, category, symptoms, root_cause, solution, and prevention to trigger cross-references and organized search across estimation, data, numerical, methodology, derivation, and replication issues.

How does YAML frontmatter pattern detection work for research documentation?

Pattern detection in research documentation works by applying consistent YAML frontmatter fields like symptoms, root_cause, and category across Markdown files, enabling agents to query and identify recurring issues across estimation, data, numerical, methodology, derivation, and replication problems.

Can I use categorized Markdown files to reproduce past research results?

Categorized Markdown files with structured YAML frontmatter enable reproduction of past research results by saving solutions under docs/solutions with metadata fields, allowing agents like learnings-researcher to accelerate subsequent analyses and replicate findings.

What's the best way to structure a documentation workflow for research issues?

The best way to structure a research documentation workflow is saving each solved problem as a single Markdown file under a category directory within docs/solutions, using YAML frontmatter fields like component, symptoms, root_cause, and solution to enable future cross-referencing and organized search.

Do I need any dependencies to create searchable research documentation?

Creating searchable research documentation requires no dependencies, only Markdown files with YAML frontmatter under category directories, using fields like component, date, category, symptoms, root_cause, solution, and prevention to build a reusable knowledge base for future sessions.