Backlog, Labels & Priorities
Stage work in a dedicated backlog before it hits the board. Tag items with custom labels and colors, rank them on a fully-customizable priority scale, and batch-promote a week's worth of work to any column in one move.
Kanban Orchestration
for
AI Coding Agents
Bring your own backlog into one board for any agent. Drag tasks to spawn sessions, see real-time status, and ship work in parallel from native terminals on your desktop.
npx kangentic To Do 2
Add user auth flow
Implement OAuth2 login with Google and GitHub providers for the dashboard.
auth feature
Refactor API error handling
Consolidate error responses to use a shared error format across all endpoints.
api
Planning 2
WS reconnection
Handle dropped connections gracefully with exponential backoff and state sync.
bug
Opus 4.6 (1M) 35%
Fix flaky integration tests
Tests intermittently fail on CI due to race conditions in database setup.
tests
Gemini 2.5 Pro 58%
Executing 2
Refactor auth middleware
Consolidate auth checks into a shared middleware and update route handlers.
auth refactor
GPT-5 Codex 72%
Generate API client types
Produce TypeScript client types from the OpenAPI schema and sync consumers.
api
Sonnet 4.5 (1M) 23%
Done 0
Drop here to complete
▾ Completed (5)
Search...
Setup CI/CD pipeline
Fix memory leak in worker pool
Add dark mode toggle
Migrate to TypeScript 5.4
Implement rate limiting
The Problem with Parallel Agents
Parallel Agents
AI coding agents can build features, fix bugs, and refactor entire modules autonomously. With git worktrees, you can run many of them in parallel and ship work at incredible speed.
Context Overload
But there's a new bottleneck: you. Five agents across three projects — which ones are stuck waiting for approval? Which one just finished? Which branch has conflicts? Juggling terminals and context-switching between agents is exhausting, and it's exactly where the speed advantage breaks down.
One Command Center
Kangentic is a command center for your coding agents. One board shows every agent's status, output, and progress. Respond when needed, and let them work autonomously the rest of the time.
Any Agent, One Board
A Kanban board for Claude Code, Codex, Gemini, Cursor, GitHub Copilot, Aider, and Warp. Spawn, suspend, and resume agent sessions with a drag.
Run Claude Code, Codex, Gemini, Aider, Cursor, GitHub Copilot, and Warp side-by-side on the same Kanban board. Drag tasks to spawn, suspend, and resume sessions. Mix agents per column and hand off context between them.
Bring your own backlog
Import issues, project cards, and work items from the tools your team already uses.
Priority Title Labels Created
⠿ Medium Setup database migrations backend 2 hours ago
⠿ None Add UI scale setting to Appearance ui 1 day ago
⠿ None Fix login redirect loop bug just now
⠿ None Add rate limiting to REST API enhancement just now
⠿ None Upgrade Webpack to v6 deps just now
Import Sources
acme/api
GitHub Issues
Add Source
GitHub Issues
acme/api Pull tasks in from GitHub Issues, GitHub Projects, or Azure DevOps — including titles, descriptions, labels, and inline images. Triage them in the backlog, then batch-promote to any column when you're ready to run agents. Already-imported items are detected automatically so re-syncing is safe.
Everything you need to orchestrate AI agents
Manage Claude Code, Codex, Gemini, Cursor, Copilot, Aider, and Warp agents from a single, intuitive interface.
Customizable Workflows
Build pipelines like Plan, Execute, Review. Set permission modes, auto-commands, and transition actions per column. Configure a plan-exit target so cards advance automatically after planning, inject prompts on column entry, and chain scripts or PRs on the way out.
Real-Time Status
See which agents are thinking or idle — right on the card, with per-agent activity detection via native hooks where available and PTY fallbacks where not. Desktop notifications fire when an agent needs your attention.
Agent-to-Board Tools
Agents that self-organize. Every running session has MCP tools to create tasks, move cards, search prior sessions, and queue follow-up work — so a planning agent can hand a backlog to an executing agent without you touching the board.
Git Worktrees & Review
Each agent runs in its own git worktree — parallel development without branch conflicts. When work is ready, the built-in Changes panel opens a split or inline diff viewer with file tree and untracked files, one click from the task card.
Session Persistence
Sessions survive restarts and crashes. Orphaned sessions are detected on startup and resumable — suspend to Done, resume later with full context, nothing is lost.
Handoff Context
Hand work between agents without losing context. When a card moves from a Claude plan column to a Codex execute column, the next agent starts with the full history of what came before.
Terminal & Activity Log
A built-in terminal for every session, plus a structured activity log that shows what each agent is doing without the noise.
Cross-Platform & Local
Runs entirely on your desktop — Windows, macOS, Linux, and WSL. No cloud service, no data leaves your machine. Uses the agent CLIs you already have installed.
Your CLIs, Your Way
No OAuth flows, no wrappers, no API proxies. Kangentic launches native Claude Code, Codex, Gemini, Cursor, Copilot, Aider, and Warp terminals. Your logins, your subscriptions or API keys. Just the real CLIs, the way each vendor intended.
How it works
Three steps from task idea to working code.
Create a Task
Add a card with a title and prompt. Paste screenshots, choose a source branch, and toggle worktree isolation — all from the create dialog.
+ New Task
✕ Add user auth flow
Write
Preview
Implement OAuth2 login with Google and GitHub providers for the dashboard.
Paste or drop files here
Priority
Medium ▾
Labels
auth feature
Branch
add-user-auth-flow-a1b2c3d4
from
main
Worktree
Auto-generated branch will be created from main in a new worktree
Cancel Create
Drag to Run
Drag the card to any active column. Kangentic creates a worktree, picks the permission mode, and spawns your chosen agent automatically.
my-project
To Do 2
#1 Add user auth flow
Implement OAuth2 login with Google and GitHub providers...
auth feature
#2 Fix dashboard layout
Sidebar overlaps main content on viewports under 1024px...
bug
Planning 1
#1 Add user auth flow
Implement OAuth2 login with Google and GitHub providers...
auth feature
Watch It Code
Your agent starts writing immediately. Follow along in the live terminal — see diffs, test results, and tool calls as they happen. Drag between columns to steer, or drag to Done to pause and pick up later.
✦ Activity
Add user auth flow
Refactor auth middleware
▾
● Update(src/auth/oauth.ts)
└ Added 3 lines, removed 1 line
- const token = getToken();
+ const token = await refreshToken();
+ if (!token) throw new AuthError();
Bash npm test -- --filter auth
✓ 12 tests passed
❯ implement the refresh logic
Frequently Asked Questions
Which AI coding agents does Kangentic work with?
Seven CLIs, all first-class:
- Claude Code (Anthropic)
- Codex CLI (OpenAI)
- Gemini CLI (Google)
- Cursor CLI
- GitHub Copilot CLI
- Aider
- Warp (Oz CLI)
Drag a task to any active column and Kangentic spawns the configured agent's CLI in an isolated git worktree, streaming the terminal into the app. Set a default agent per project and override it per column, so a task can hand off from Claude in Planning to Codex in Executing without leaving the board.
Can I mix multiple agents on the same board?
Yes. Set a default agent per project and override it per column, so a task can move through Planning with Claude, Executing with Codex, and Code Review with Gemini on the same board. When a column has handoff context enabled, Kangentic passes the prior agent's native session history file to the next agent, so conversation context carries across even when the agent changes.
Do I need an API key or cloud account to use Kangentic?
No. Kangentic uses the agent CLIs you already have installed with the credentials you have already configured. There are no OAuth flows, no API proxies, and no accounts. All data stays on your machine in a local SQLite database.
Is Kangentic free and open source?
Yes. Kangentic is free forever under the AGPL-3.0 license, with every feature included and no paid tiers. The source is at github.com/Kangentic/kangentic.
What operating systems does Kangentic support?
Kangentic is a cross-platform Electron desktop app with native installers for Windows (NSIS), macOS (DMG), and Linux (deb, rpm). WSL users should install the Windows version.
How is Kangentic different from running agents in separate terminals?
Running multiple agents in separate terminal tabs scales poorly: you context-switch between windows, lose track of what each agent is doing, and have no shared view of progress. Kangentic replaces that with a Kanban board where every column transition controls the agent lifecycle automatically, each task gets its own git worktree, and a unified activity view shows which agents are thinking, idle, or waiting for approval.
Built with
Modern, battle-tested technologies for reliability and performance.
Electron
xterm.js
SQLite
React
TypeScript
Tailwind
kangentic.com
❯ claude "Build the landing page for kangentic.com"
✓ Created src/pages/index.astro
✓ Added hero section with animated Kanban mockup
✓ Built 3-step "How It Works" flow
✓ Deployed to Cloudflare Pages
This site was built with Kangentic.▊
Yes, really. Agents built this whole site, orchestrated through Kangentic.
Ready to orchestrate your agents?
Get started in under a minute. Free and open source.
Get Started