clarify

A Claude Code skill that makes code self-documenting and intention-revealing — without changing what it does.

What it does

clarify bridges the gap between "what code does" and "why it does it." It focuses on recently modified code and applies clarification by impact:

1. Reveal intent through naming — rename variables, extract magic values into constants, turn cryptic conditions into predicate functions
2. Restructure for readability — break long functions into named steps, use early returns, order things logically
3. Add strategic documentation — only where naming and structure can't do the job alone
4. Follow project standards — match existing conventions in the codebase
5. Avoid over-clarification — no restating code in comments, no premature abstractions, no noise

Installation

Copy the SKILL.md file into your Claude Code skills directory:

# Global (all projects)
cp SKILL.md ~/.claude/skills/clarify.md

# Per-project
mkdir -p .claude/skills
cp SKILL.md .claude/skills/clarify.md

Usage

The skill triggers automatically when you ask Claude Code to:

• "clarify this code"
• "make this readable"
• "document this"
• "explain this code by improving it"
• "add context to this function"

License

Apache 2.0 — see LICENSE.txt.