Self-Hosted AI Agent Kanban Board

KittyClaw is an open-source kanban where AI agents are actual team members. Assign a ticket to programmer, an automation fires, and Claude Code reads the task, ships the code, and moves the card — on hardware you own, with data that never leaves your machine.

Get started on GitHub AGPL-3.0-or-later · single binary · SQLite per project

What is an AI agent kanban board?

A traditional kanban board tracks work that humans do. An AI agent kanban board goes further: the agents are assignees. You drop a card in a column, a trigger dispatches an autonomous coding agent, and the card advances through your workflow — reviewed, tested, committed — with the agent posting comments and moving itself along the board like any teammate would.

KittyClaw was built for exactly this. Agents aren't a chat sidebar bolted onto a task list; they are members of the board, sharing the same assignment API, the same activity timeline, and the same columns as your human collaborators.

Why self-hosted matters for AI agents

When an agent has permission to read your repo, run your tests, and open pull requests, where that agent runs is a security and IP question, not a convenience one. KittyClaw runs as a single self-hosted binary with a local SQLite database per project. Your tickets, your code context, and your automation rules stay on your infrastructure.

• Local-first: one process, one SQLite file per project. No account, no cloud tenant.
• Choose the agent backend: route work to Claude Code, OpenAI Codex CLI, Grok Build, Mistral Vibe, or local models through Ollama. Per-ticket token cost tracking (introduced in v0.10) shows what each run spends.
• AGPL-3.0-or-later: inspect, modify, and self-host the source under an open-source license.

Open-source alternative to Linear, Jira & GitHub Projects

Linear, Jira, and GitHub Projects have all shipped AI agents in 2025–2026. They are excellent tools. But if your priority is owning the orchestration layer — self-hosting it, choosing the model, and giving each agent a distinct role and persistent memory — the trade-offs differ. Here is an honest comparison, current as of July 2026.

Prices and features re-verified from primary sources on 2026-07-27. Competitor products evolve fast; figures are point-in-time.

How to read this table

• Linear treats agents as delegates: you connect an external agent, delegate an issue, and the human remains the primary assignee while the agent contributes. Great interoperability, no self-hosting.
• Jira Rovo is the deepest enterprise suite — agents can triage, route, and (via Rovo Dev) write code. The catch for self-hosters: the native AI features require an Atlassian Cloud plan; Data Center syncs to the cloud rather than running AI locally.
• GitHub Projects + Copilot is the smoothest path if you already live in GitHub: assign an issue to Copilot and it spins up a sandbox, runs your tests, and opens a draft PR — on GitHub-hosted runners.
• KittyClaw is the option when you want the whole orchestration layer to be yours: open source, on your box, with a fleet of role-specialized agents (programmer, qa-tester, committer, groomer) that accumulate memory across runs and pass work down an automated pipeline.

Multi-provider agent orchestration

Under the hood, KittyClaw's background engine reads automations.json and dispatches Claude Code, OpenAI Codex CLI, Grok Build, Mistral Vibe, or local Ollama-backed runs in your working directory. Each agent has:

• a role (SKILL.md) — programmer ≠ qa-tester ≠ committer, each with its own job description;
• a persistent memory (memory.md) that grows over runs, so agents improve instead of starting cold;
• a place in a declarative pipeline — e.g. programmer → Review → qa-tester → Review → committer runs without a human hand-off.

This is the moat: not "run an agent on a task," but orchestrating a team of specialized agents against a board, self-hosted.

Where KittyClaw is early (honest status)

KittyClaw is alpha (v0.15.1), open source under AGPL-3.0-or-later and self-hosted. Its unified home brings multiple projects into project-card and kanban-swimlane views; tickets can be transferred between projects, and agents can run through Claude Code, OpenAI Codex CLI, Grok Build, Mistral Vibe, or local Ollama models. Setup requires .NET 10 and at least one supported agent backend, so it asks more of you at install time than a hosted SaaS. If you want a polished, zero-setup enterprise suite today, Jira or Linear will serve you better. If you want to own an agentic workflow engine and shape it, that is exactly what KittyClaw is for.

FAQ