mercurialsolo

mercurialsolo / claudectl

Public

Mission control for Claude Code. Approve, catch blockers, control spend, and kill claude code sessions - without tab hunting.

agent-ops ai-agents claude claude-code cost-tracking
24
2
100% credibility
Found Apr 15, 2026 at 28 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

claudectl is a terminal dashboard for supervising, budgeting, recording, and coordinating multiple Claude Code AI coding sessions.

How It Works

1
🚀 Discover claudectl

You hear about a handy dashboard that watches your AI coding helpers and makes teamwork easy.

2
📥 Start with one click

Download and launch the colorful screen that shows all your AI sessions at once.

3
👀 See everything clearly

Spot which helpers need your okay, track spending, and catch issues like stalls or high costs instantly.

4
🤖 Add smart auto-pilot

Connect a local thinker to suggest safe approvals or blocks, learning from your choices.

5
🎥 Record magic moments

Press one key to save a quick video highlight of file changes and commands.

6
🔗 Team up sessions

Run linked tasks so your AIs pass results to each other automatically.

✅ Ship code faster

Your projects build quicker with AI sessions under easy control—no more tab hunting.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 28 to 24 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 claudectl?

claudectl is a Rust-built TUI mission control center for Claude Code sessions, letting you supervise multiple AI coding agents from one dashboard without tab-hunting. It monitors session health for cache issues, cost spikes, loops, stalls, and context saturation, while enforcing budgets, detecting file conflicts, and auto-killing blockers. Press one key to approve prompts, record highlight-reel GIFs, or orchestrate dependency-ordered workflows via JSON tasks.

Why is it gaining traction?

Unlike bare Claude Code or GitHub Copilot workspaces, claudectl adds a local LLM brain (ollama, llama.cpp) for auto-approving safe tools like reads or tests while denying destructives like rm -rf. Developers love the live $/hr burn rates, desktop notifications on needs-input, and cross-terminal switching for iTerm2, Kitty, or WezTerm. Zero-config startup under 50ms and Homebrew install make it a mission control shortcut for scaling AI agents.

Who should use this?

Backend devs juggling Claude Code for API refactors across monorepos, ML engineers spawning parallel training pipelines, or infra teams running terraform sessions. Ideal for anyone hitting spend limits or file conflicts in multi-agent workflows on macOS or Linux.

Verdict

Try claudectl if you're deep into Claude Code—its feature density punches above 24 stars and 1.0% credibility score. Early maturity means watch for edge cases, but MIT license, full docs, and demo mode lower the risk.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.