coderabbit-review
Handle CodeRabbit AI review cycles after creating a PR
Think through a problem on paper before acting
$ npx -y skills add markmdev/meridian --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
/thinkContext preview
What this command does when you run it.
Think through a problem on paper before acting
description: Think through a problem on paper before acting argument-hint: [LINES] [TOPIC]
You have been asked to think through a problem before acting.
**Arguments:** $ARGUMENTS
Parse the arguments:
**Instructions:**
1. Create a scratch file: `.meridian/.scratch/thinking-YYYY-MM-DD-HH-MM-topic.md`
2. Write your thoughts to this file. You have **{line_limit} lines** to think.
3. End with a brief "## Synthesis" section summarizing:
4. After writing the thinking file, tell the user what you figured out and ask how to proceed.
**The point:** Thinking on paper catches issues early and produces better solutions than jumping straight to action. Use the line limit — don't stop early, don't exceed it significantly.
Meridian makes Claude Code more reliable on real projects. It adds persistent project context, smarter session handoff, and lightweight workflow enforcement so Claude is less likely to lose the plot halfway through a long task.
Handle CodeRabbit AI review cycles after creating a PR
Start a work-until loop that keeps you working on a task until completion