millionco

millionco / claude-doctor

Public

Diagnose your Claude Code sessions

claude code doctor
227
14
100% credibility
Found Apr 16, 2026 at 227 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

claude-doctor examines chat histories from AI coding sessions to identify common pitfalls and generate personalized rules for better performance.

How It Works

1
💭 Frustrated with AI chats

You've noticed your AI helper keeps making the same mistakes or getting stuck in loops during projects.

2
🔍 Discover claude-doctor

You find this friendly checker that scans your past chat histories to spot patterns and suggest fixes.

3
🚀 Run the checker

With one quick command, it scans all your project chats or just the latest one you pick.

4
📊 See the health report

A clear report pops up showing problem spots like too many restarts, frustrated messages, or edit loops, with a health score and timeline.

5
✨ Get smart rules

It creates ready-to-use rules tailored to your habits, like 'plan all changes before editing' or 'change approach after errors'.

✅ Better AI teamwork

Paste the rules into your guide file, and your next sessions run smoother with fewer frustrations and better results.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 227 to 227 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is claude-doctor?

Claude-doctor is a TypeScript CLI for diagnosing Claude code sessions by scanning transcripts in ~/.claude/ for anti-patterns like edit thrashing, error loops, user frustration, and excessive file reads. Run `claude-doctor` to analyze all sessions, `claude-doctor ` for specifics, or `--rules` to generate paste-ready fixes for CLAUDE.md/AGENTS.md—solving claude doctor hangs, claude doctor not working, claude doctor stuck, and similar session failures. It outputs health scores, timelines, and JSON for scripting claude diagnose workflows.

Why is it gaining traction?

With 227 stars, it hooks Claude-heavy devs by turning raw session logs into actionable insights no generic tool matches—spotting claude doctor prompt drifts or tool inefficiencies instantly. The `--save` flag builds a project model with baselines, while signals like correction-heavy or keep-going-loop reveal why sessions restart. Simple npm install or npx skips claude doctor zsh command not found claude issues.

Who should use this?

Claude Code users debugging failed sessions, prompt engineers tuning AGENTS.md rules, or teams fixing claude doctor auto update failed and claude doctor config install method unknown errors. Ideal for devs analyzing tf diagnose github-style logs across projects.

Verdict

Grab it if you're deep in Claude sessions—solid README and CLI make it instantly useful despite 1.0% credibility score and modest stars signaling early maturity. Test coverage looks basic; watch for claude doctor update stability before production pipelines.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.