Kangentic documentation

Kangentic is an agentic development environment (ADE): a cross-platform desktop app where a Kanban board runs your AI coding agents. It runs fourteen coding-agent CLIs, Claude Code, Codex, Gemini, Cursor CLI, GitHub Copilot CLI, Grok Build, Antigravity CLI, Goose CLI, OpenCode, Aider, Qwen Code, Kimi Code, Factory Droid, and Warp (Oz CLI), plus Ollama for local models. Instead of managing agents one at a time in separate terminals, you spawn, monitor, suspend, and resume them all from one board, each task isolated in its own git worktree.

Start here

Quickstart Install, open a project, and run your first task.

Installation Installers per platform, npx, and building from source.

Creating tasks Task descriptions, image attachments, and column management.

Why Kangentic?

A coding agent can carry a task on its own, and git worktrees let several run at once without touching each other’s files. The limit is then you: which agent is doing what, which one is waiting on a permission prompt, and where each branch went. Kangentic puts every session on one board, shows the state of each card, and flags the ones that need you.

It runs on your desktop with the agent CLIs and credentials you already have. Your boards, tasks, and transcripts stay in a local database, never on a Kangentic server. See the Privacy Policy.

How it works

1. Create a task - Add a card to your board describing the work (bug fix, feature, refactor)
2. Assign and run - Drag the card to the Executing column to spawn the column’s agent
3. Monitor progress - Watch the agent work in a real-time terminal, or let it run in the background
4. Review and merge - When the agent finishes, review the diff in the built-in Changes panel and merge

Each agent session runs in its own git worktree with a dedicated terminal, so you can run many agents in parallel without interference.

What you bring

Kangentic is deliberately unopinionated about three things. The agents you run, the work you feed it, and the process you run that work through are all yours.

Any agent, one board

Pick a default agent per project, override it per column, and hand work between agents with their context intact. Sessions survive crashes and restarts, and Kangentic reads each agent’s own activity signals so the board can tell you which ones are thinking, which are idle, and which are waiting on you.

• Agent orchestration - the supported CLIs, per-column overrides, and cross-agent handoff
• Activity detection - how the board knows what each agent is doing
• Session persistence - what survives a restart, and what resuming does

Bring your own backlog

Stage work before it reaches the board, with labels, priorities, and filtering, then promote a week’s worth to any column in one move. Import from GitHub Issues, GitHub Projects, Azure DevOps, and Asana; re-syncing is safe because already-imported items are detected automatically.

• Backlog and imports - staging, labels, priorities, bulk operations, and importing

Bring your own workflow

The seven columns Kangentic seeds are a starting point, not the product. Rebuild them: each column carries its own agent, model, reasoning effort, permission mode, spawn behavior, handoff, session target, plan-exit target, and automations: a message to the agent, a move to another column, a script, a webhook, or a notification, when a task enters, when its agent finishes a turn, or when it leaves. Share the result with your team through kangentic.json, and keep personal deviations in kangentic.local.json.

A workflow goes one step further: a set of linked tasks, each with its own route of columns, that starts each task when the ones it waits for reach Done and moves it on as each column’s work finishes.

The workflow runs both ways. Every spawned session holds the board’s own MCP tools, so an agent can search past conversations, message another task’s agent, and move its own card when the work is done. The same history is yours to question: the Knowledge Graph maps every past conversation by meaning and answers questions about the work through the agent you choose.

• Columns - what a move does to the agent, and the Column Manager
• Automations - what a column runs as a task enters, at each turn end, and as it leaves
• Workflows - linked tasks that run through the board on their own
• Board configuration - sharing a board layout through kangentic.json
• Board profiles - named alternate ladders of per-column settings
• How the MCP server works - how the board’s own tools reach a running agent, with all 90 of them in the MCP tool reference

Where things are

The docs follow the sidebar, and each group answers one kind of question. Inside a group the pages run from what you need in the first hour to what waits until you change a default; a page marked Advanced in the sidebar can wait.

• Getting started - the first hour: what Kangentic is, installing, first launch, and the first project.
• Tasks and columns - the board: task cards, the Backlog and imports, and what columns are and do, including automations, workflows, board profiles, and kangentic.json.
• Agents and sessions - the process behind a card: which CLIs run, worktrees and pull requests, what a session stores, how activity is detected, and conversation memory.
• Windows and panes - the task detail window and the panes inside it: terminal and Activity tab, Changes, the embedded browser, the Command Terminal, and docking and pop-out.
• Across projects - the surfaces outside any one board: the project sidebar and project groups, Quick Find, Agent Monitor, Usage Stats, the Knowledge Graph, and notifications.
• Settings and configuration - the settings panel, the configuration reference, hotkeys, command shortcuts, and dictation.
• MCP Server - how the board’s own tools reach a running agent, all 90 of them in the MCP tool reference, and the trust boundary.
• Help and privacy - the app as a whole: troubleshooting, auto-updates, what leaves the machine, and the glossary.
• Kangentic Mobile - the phone companion: what it can do, installing, pairing, push notifications, and encryption.
• Relay server - the server behind pairing: the hosted relay, self-hosting, the reference, and the security model.
• Contributing and internals - building, testing, packaging, the architecture, and shell support, for people changing Kangentic rather than using it.

Explore further

Docking and pop-out windows The window layer: floating, tiled, snapped, docking by pointer, and the surfaces that pop out to a second monitor.

Settings 12 themes, per-column customization, and permission modes.

Worktrees and pull requests Branches, the base branch, PR detection, and what happens to a worktree at each move.

Knowledge Graph A map of your past agent conversations by meaning, and questions answered from your tasks, conversations, and code.

Reviewing changes The built-in diff viewer, commit graph, and viewed marks.

Kangentic Mobile Check on agents, answer prompts, and review diffs from your phone.

Kangentic Relay The blind rendezvous server behind mobile pairing, and how to run your own.

Troubleshooting Common problems, log locations, and how to recover a stuck board.