skills-for-humanity: Instructions file for Claude Code

skills-for-humanity

A structured thinking toolkit delivered as Claude Code skills. Each skill encodes a complete reasoning methodology — not just instructions, but the full technique: when to apply it, when not to, how to sequence the steps, and what rigorous output looks like.

Structure

skills/
  {category}/           ← category command, invoked as /{category}
    SKILL.md
  {category}-{skill}/   ← individual skill, invoked as /{category}-{skill}
    SKILL.md

docs/
  {category}/
    README.md           ← category index with skills table
    {skill}.md          ← full skill documentation with example

Naming convention

• Category commands: skills/{category}/SKILL.md → invoked as /{category}
• Individual skills: skills/{category}-{skill}/SKILL.md → invoked as /{category}-{skill}

Example: skills/ethics-council/SKILL.md is invoked as /ethics-council. The folder name IS the command name.

SKILL.md format

Every SKILL.md starts with YAML frontmatter:

---
name: {command-name}
description: "One or two sentences describing what this skill does and what triggers it. Include the natural-language phrases that should activate it."
---

The body contains the full methodology: context/rationale, step-by-step process, output format, and notes on when to use vs. adjacent skills.

Category command format

Category commands (skills/{category}/SKILL.md) diagnose the situation and route to the right individual skill. They read the user's input, identify which sub-skill fits, and invoke it inline. Pattern: describe the situation → diagnose the problem type → apply the right tool.

Docs format

Each skill documentation page (docs/{category}/{skill}.md) follows this structure:

1. Title + tagline
2. What it does (2–3 sentences)
3. When to use it (bullet list)
4. When NOT to use it (2 paragraphs with the nearest-neighbor trap)
5. Full realistic example with input and complete structured output
6. Related skills (cross-links)