ClaudeMap

Vibe Kanban

Kanban orchestration for 10+ AI coding agents — plan, run, and review their work on one board.

Plugins & Agentsorchestrationkanbanmulti-agentparallel

Vibe Kanban turns agent coding into a kanban workflow: stack tasks, assign them to Claude Code, Codex, Gemini CLI, Cursor, and other agents, review diffs, and merge — all from one board. Note: the hosted cloud service was sunset in 2026 (announced April 10); the project now ships as fully local architecture, open source and community-maintained.

What it does

  • Run multiple coding agents in parallel across a kanban board with status tracking
  • Agent-agnostic: Claude Code, Codex, Gemini CLI, GitHub Copilot, Amp, Cursor, OpenCode, Droid, CCR, Qwen Code
  • Review-first loop: inspect diffs in the browser before anything lands on a branch
  • Fully local architecture after the 2026 cloud sunset — self-host with npx or Docker

When to use it

Best for solo devs or small teams running 2-4 agent tasks in parallel without a full CI-style pipeline. Skip it if you need managed multi-user project features — those died with the cloud service.

Getting started

npx vibe-kanban

# or Docker for a persistent self-hosted instance (see the repo deployment guide)

Things to know

Cloud features (issues, comments, projects, organisations) were dropped in the sunset; the community roadmap is still taking shape, so expect project-management gaps. A data export feature ships in the latest version.

Repository

BloopAI/vibe-kanban

Charted

Related on the map

  • clawd-conduit

    Plugin / Agent

    Wire all 15 Claude Code hook events into the Clawd on Desk desktop pet — and turn permission requests into a blocking desktop card.

    • hooks
    • desktop
    • notifications
    • claude-code
  • CometixCode

    Plugin / Agent

    An unofficial Rust 1:1 reimplementation of the Claude Code terminal UI, ported file-for-file onto an iocraft TUI framework.

    • terminal-ui
    • rust
    • reimplementation
    • unofficial
  • Abide

    Plugin / Agent

    Hook-level enforcement for the natural-language rules in AGENTS.md and CLAUDE.md — every edit is judged and violations are repaired in the same turn.

    • hooks
    • code-review
    • rules
    • agents-md