Overview
Vibe Kanban is an open-source tool for managing the work that AI coding agents do. You plan tasks as cards on a kanban board, then hand each task to a coding agent that runs in its own workspace with a branch, a terminal, and a dev server. The idea is that engineers now spend most of their time planning and reviewing agent output, so the fastest way to ship more is to make planning and review smoother.
Each workspace keeps an agent's work isolated, so you can review the resulting code diff, leave inline comments, and send feedback to the agent without leaving the UI. A built-in browser with devtools lets you preview the running app. When the work looks good, you can open a pull request with an AI-generated description and merge it. Vibe Kanban works with more than ten coding agents, including Claude Code, Codex, Gemini CLI, GitHub Copilot, Amp, Cursor, OpenCode, Droid, CCR, and Qwen Code.
Note: the project maintainers have announced that Vibe Kanban is sunsetting.
What it does
- Plan and prioritise work as issues on a kanban board, privately or with a team
- Run coding agents in isolated workspaces, each with its own branch, terminal, and dev server
- Review code diffs and leave inline comments that go straight back to the agent
- Preview the running app in a built-in browser with devtools, inspect mode, and device emulation
- Switch between 10+ coding agents including Claude Code, Codex, Gemini CLI, GitHub Copilot, Amp, and Cursor
- Open pull requests with AI-generated descriptions, review on GitHub, and merge
Getting started
Vibe Kanban runs from a single command. First authenticate with the coding agent you want to use, then start the app and plan your first task.
Authenticate your coding agent
Make sure you have logged in to your preferred coding agent (for example Claude Code, Codex, or Gemini CLI). The docs list every supported agent and how to set each one up.
Launch Vibe Kanban
Run the app straight from npm with a single command. It starts the board in your browser so you can create your first issue.
npx vibe-kanbanPlan, run, and review
Create issues on the kanban board, open a workspace to let an agent execute the task on its own branch, then review the diff, leave inline comments, and open a pull request when it is ready.
Commands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Planning a backlog of issues and assigning each one to an AI coding agent to implement
- Running several coding agents in parallel, each in its own isolated workspace and branch
- Reviewing agent-generated diffs with inline comments before opening a pull request
- Previewing a running app in a built-in browser to test agent changes before merging
How Vibe Kanban compares
Vibe Kanban alongside other open-source autonomous coding agents tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| opencode | ★ 176k | OpenCode is an open source AI coding agent that runs in your terminal, with built-in build and plan agents and an optional desktop app. |
| Gemini CLI | ★ 105k | An open-source command-line AI agent from Google that connects your terminal to Gemini models for reading code, editing files, running shell commands, and searching the web. |
| OpenHands | ★ 77.8k | An open-source AI software-development agent that plans tasks, edits files, runs commands, and tests code, usable from a terminal CLI, a local web GUI, or a Python SDK. |
| Goose | ★ 49.8k | Goose is an open-source AI agent that runs on your machine for code, research, automation, and more, with a desktop app, CLI, and API built in Rust. |
| Continue | ★ 34.1k | Continue is an open-source coding agent available as a CLI, VS Code extension, and JetBrains plugin, plus version-controlled AI checks that run on pull requests in CI. |
| Vibe Kanban | ★ 27.1k | A kanban board to plan, run, and review work across many coding agents |
| Serena | ★ 25.6k | Serena is an open-source MCP toolkit that gives any LLM coding agent symbol-level code retrieval, editing, and refactoring across 40+ languages. |
| Crush | ★ 25.5k | Crush is a terminal coding agent from Charm. It connects your code, tools, and workflows to any OpenAI- or Anthropic-compatible LLM, with LSP context and MCP plugins. |