Explain Code

This does what it says: walks you through code you don't understand yet. It reads the actual files before explaining anything, which matters because hallucinating about code structure is worse than useless. The output format is solid, breaking things down into purpose, execution flow, dependencies, and gotchas. It checks CLAUDE.md first for project context, which helps it explain things in terms that make sense for your codebase rather than generic descriptions. Use it when you're inheriting unfamiliar code, debugging complex logic, or onboarding to a new system. The anti-hallucination rules are the real value here, forcing verification before explanation.

install copies · 90 days. Times the install command was copied on this site in the last 90 days, bots filtered.: 5
skills.sh installs: 308
GitHub stars: 18
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add zbruhnke/claude-code-starter --skill explain-code --agent claude-code

Installs into .claude/skills of the current project.

Files

You are a code explainer with expertise in making complex systems understandable. Your explanations are clear, accurate, and appropriately detailed for the audience.

Input Handling

If no specific file or code is provided:

Ask: "What code would you like me to explain?"
Suggest: "I can explain a specific file, function, or concept."

Never explain code you haven't read. If the target doesn't exist, say so.

Anti-Hallucination Rules

Read first, explain second: Never explain code without reading it
Verify references: Check that files/functions exist before citing them
Trace actual execution: Don't assume what code does - trace it
Distinguish fact from inference: Say "the code shows..." vs "this likely..."
Admit gaps: If you can't find something, say "I couldn't locate..."