dukex

dukex / crewbit

Public

Give life to an AI agent with a single command.

11
0
100% credibility
Found Mar 31, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Crewbit monitors issue trackers like Jira or GitHub Projects and automatically triggers AI coding sessions via Claude Code slash commands on prioritized tasks defined in a YAML workflow file.

How It Works

1
🔍 Discover crewbit

You hear about a handy tool that automates your development tasks using AI to handle issues from your task list.

2
📥 Install easily

Grab the tool with a simple download command, and it's ready on your computer in moments.

3
📝 Create your plan

Write a short list of instructions telling the AI what kinds of tasks to pick up and how to tackle them, like starting new work or finishing reviews.

4
🔗 Connect your tasks

Link it to your Jira board or GitHub project so it can see your ready-to-go issues.

5
▶️ Start your helper

Turn on the always-watching assistant, and it begins scanning for work automatically.

6
👀 Watch it go

Relax as it grabs the highest priority task, lets the AI think and code on it safely, then moves to the next one.

🎉 Work gets done

Your task list clears up on its own, with the AI handling the routine coding so you focus on the big picture.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 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 crewbit?

Crewbit is a TypeScript CLI tool that launches persistent AI agents to automate dev workflows from issue trackers like Jira or GitHub Projects. Point it at a YAML config with `crewbit ./dev-junior.yaml`, and it polls for ready issues in priority order, spawning isolated git worktrees where Claude Code slash commands like `/develop` or `/merge` handle the work. It solves the drudgery of manually feeding tickets to AI, giving life to agents that implement features, run releases, or QA tests autonomously.

Why is it gaining traction?

Unlike one-off AI prompts, crewbit daemonizes the loop—watching queues, retrying failures, and cleaning up worktrees— all defined in one YAML file with ready examples for junior devs or copywriters. The hook is simplicity: curl install, env vars for auth like GITHUB_TOKEN or JIRA_API_TOKEN to give repo access, and dry-run flags to test without chaos. It skips permission prompts via Claude's dangerous flag, trading safety for speed on trusted machines.

Who should use this?

Indie hackers or small teams on Jira/GitHub Projects who want AI to chew through backlogs—think solo founders auto-implementing "Ready" tickets, releasers handling merges on "Accepted," or translators batching i18n. Ideal if you're already using Claude Code and need to give GitHub repo read access to private issues without constant babysitting.

Verdict

Try it for prototyping AI-driven workflows if you review slash commands and run on isolated envs—11 stars and 1.0% credibility scream early alpha, but solid YAML docs and tests make it playable. Skip for production until more battle scars.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.