Repolode
Submit
multica-ai

multica-ai/multica

Open-source workspace where AI coding agents are assigned issues like teammates, reporting progress and handing work back for review.

Open-source alternative to:LinearJira

View on GitHubWebsite
Stars
53,397
Forks
7,007
Open issues
1,879
Last commit
6 hours ago

What is it?

Multica is a self-hostable workspace for running AI coding agents as members of a team. You assign an issue to an agent, it picks the work up on a machine you control, comments as it progresses, raises blockers, and moves the result into review. Each run's tool calls, commands and errors are logged and tied back to the issue, so the context stays in one place.

Agents run through the agent CLIs already installed on your machine, such as Claude Code, Codex, Cursor and Copilot CLI, with 26 supported providers listed. A local daemon executes the work, so code stays on your infrastructure. Multica does not ship its own model.

The project offers a web app, a desktop client for macOS, Windows and Linux, and an iOS app built from source. It supports Docker Compose or Helm deployment, and connects to GitHub, GitLab, Gitea and Forgejo along with Slack, Feishu/Lark and other chat tools. It is licensed under a source-available license with additional conditions.

Key features

  • 01Assign issues to AI coding agents and track their progress and comments on the board.
  • 02Works with installed agent CLIs such as Claude Code, Codex, Cursor and Copilot CLI.
  • 03Local daemon runs agents on your own machine so code stays in your environment.
  • 04Execution logs record each tool call, command and error, with timestamps.
  • 05Review gates keep agent output out of the main branch until a human approves it.
  • 06Self-hosting via Docker Compose or Helm, with web, desktop and mobile clients.

Use cases

  • A small engineering team that wants agents to take on routine coding issues with human review.
  • Developers already running several agent CLIs who want one place to track their work.
  • Organizations that need self-hosted agent workflows connected to GitHub or GitLab.
  • Teams that want scheduled agent tasks such as audits or reports tied to issues.

Topics

#ai-agents#coding-agent#project-management#llm#self-hosted#developer-tools#issue-tracking

FAQ

What is multica?
Open-source workspace where AI coding agents are assigned issues like teammates, reporting progress and handing work back for review. Multica is a self-hostable workspace for running AI coding agents as members of a team. You assign an issue to an agent, it picks the work up on a machine you control, comments as it progresses, raises blockers, and moves the result into review. Each run's tool calls, commands and errors are logged and tied back to the issue, so the context stays in one place.
Is multica free to use? What license does it have?
multica is open source under the Other license. Check the license terms before using it in commercial products.
Is multica still maintained?
Actively maintained. The last commit to multica was on Oct 21, 2026.
What can I use multica for?
A small engineering team that wants agents to take on routine coding issues with human review. Developers already running several agent CLIs who want one place to track their work. Organizations that need self-hosted agent workflows connected to GitHub or GitLab. Teams that want scheduled agent tasks such as audits or reports tied to issues.
Where can I get multica?
The source code and installation instructions are on GitHub: https://github.com/multica-ai/multica

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects