Suits developers and small teams already using AI coding agent CLIs who want to manage them together; you should have at least one agent CLI installed and signed in first.

What can you build with multica?

Assign a bug report to Claude Code or Codex and review the pull request it produces. Run several different coding agents on one board and compare their runs and costs. Schedule a weekly audit or report with an autopilot so an agent does it without a reminder. Self-host an agent workspace so code stays on your own machines. Trigger and follow agent work from Slack or another chat tool your team already uses.

multica-ai /

multica

A shared board where you assign issues to AI coding agents like teammates

53.6k stars
1.8k stars since 3 Oct
Go
Updated today
Some experience
AI agents

Open on GitHub (opens in a new tab) Website (opens in a new tab)

If you use several coding agents, each one lives in its own terminal tab and forgets context when the session ends. Multica puts those agents and your human teammates on one board. You file an issue, set an agent as the assignee, and it picks the work up, comments as it goes, and moves the issue to review when done.

It does not ship a model. A small daemon runs on your laptop or a cloud machine, next to your code, and launches the agent CLIs you already have installed and signed in. The README lists 26 of them, including Claude Code, Codex, Cursor Agent, Copilot CLI and Kimi. A Go backend with PostgreSQL keeps the issues, runs and logs; web, desktop and mobile apps sit on top.

The idea is that the intent, the run, the decisions and the diff all stay attached to one issue. Work lands in review rather than in main, so a person decides what ships. Note the README calls it source-available, and the license field on GitHub is not a standard one, so check the terms before you build on it.

What it does

Agents as teammates
Give each agent a name, a provider and a runtime, and it appears on the board like a person. Squads let agents and people share a team, with a leader routing the work.
Works with agent CLIs you have
Drives 26 CLIs, such as Claude Code, Codex, Cursor Agent and OpenCode. Switching provider is a dropdown choice.
Execution log and cost tracking
Replay every tool call, command and error with timestamps, and see what each run cost per agent and per issue.
Steering, retries and review gates
You can reply to a running agent and the message lands in the current run (Claude Code, Codex and Grok today). Failed runs retry or stop with a reason, and work goes to review first.
Autopilots and self-hosting
Schedule recurring jobs like standups and audits on a cron. Run the whole stack with Docker Compose or Helm, and connect GitHub, GitLab, Gitea or Forgejo.

What you could build

Assign a bug report to Claude Code or Codex and review the pull request it produces.
Run several different coding agents on one board and compare their runs and costs.
Schedule a weekly audit or report with an autopilot so an agent does it without a reminder.
Self-host an agent workspace so code stays on your own machines.
Trigger and follow agent work from Slack or another chat tool your team already uses.

Get started

Easiest is the cloud sign-up at multica.ai or the desktop app, which registers your computer as a runtime. Make sure an agent CLI like Claude Code is installed and signed in, then create an agent in the sidebar, file an issue and assign it. To self-host, you need Docker and the two commands below.

curl -fsSL https://raw.githubusercontent.com/multica-ai/multica/main/scripts/install.sh | bash -s -- --with-server

multica setup self-host

Commands are copied from the project's README. Read it on GitHub for the full setup.