wbopan

wbopan / cotask

Public

Easy parallel project management for Claude Code agents.

17
2
100% credibility
Found Mar 23, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Cotask is a dashboard and set of commands for managing tasks across multiple Claude Code AI agents using a shared markdown file in your project.

How It Works

1
🔍 Discover Cotask

You hear about Cotask, a handy tool that helps organize tasks for your AI coding helpers so they work together smoothly on projects.

2
📥 Add Cotask helper

In your AI coding app, you use a simple command to bring Cotask into your projects, making task management easy.

3
Create your first task

Tell your AI to make a new task with a quick note about what needs doing, and it appears in a shared project list.

4
🖥️ Open the dashboard

Your AI launches a colorful web page in your browser showing all tasks with drag-and-drop lists for backlog, to-do, ongoing, and done.

5
🔄 Start work on tasks

Drag tasks to 'ongoing' and tell your AI to tackle one by its name, watching it plan and execute right away.

6
👀 Track progress live

The dashboard updates in real time with task statuses like running or waiting, plus buttons to jump straight to active work screens.

🎉 Master parallel projects

Now you effortlessly manage multiple AI helpers on different tasks, staying in control with clear overviews and easy interventions.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 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 cotask?

Cotask is a JavaScript plugin for Claude Code agents, built on Node.js, that streamlines parallel project management via a shared TASKS.md file. Developers get a local web dashboard for drag-and-drop task editing, real-time monitoring of agent sessions (running, idle, permission waits), and one-click jumps to terminals—perfect for easy parallel workflows like parking multiple agents on subtasks. Commands like `/dashboard`, `/creating-task `, and `/starting-task #slug` handle setup, with tasks tracking lifecycle via checkboxes ([ ] todo, [/] ongoing, [x] done, [-] backlog) plus acceptance criteria and completion memos.

Why is it gaining traction?

It stands out by keeping humans in the loop with a visual dashboard suited for oversight, while agents parse structured markdown—no clunky APIs or shared state hacks. The real-time session linking and Ghostty integration cut cognitive load for multi-agent runs, making parallel circuit-style coordination feel effortless, unlike basic task lists. Early adopters praise the demo's intuitive flow for easy GitHub projects to contribute.

Who should use this?

Claude Code users on macOS juggling multiple AI sessions for large codebases, like backend devs splitting refactors or full-stack teams on feature branches. Ideal for those tired of terminal tab chaos in agent-driven dev, especially with Node.js 18+ setups seeking cotasker app simplicity over full PM tools.

Verdict

Try it if you're deep into Claude Code—solid docs and demo make onboarding easy despite 17 stars and 1.0% credibility signaling early maturity. Lacks broad testing, but low-risk for personal projects; watch for cross-platform growth.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.