Multica

Run several coding agents and each ends up in its own terminal tab, forgetting context between sessions and demanding the same explanation four times a day. Multica puts those agents and the people they work with on one board, where an agent takes an issue, works it, comments as it goes and hands the result back for review.

Multica ships no model of its own. It drives agent CLIs already installed on a machine you control, 26 of them including Claude Code, Codex, Cursor, Copilot, Kimi and OpenCode, and the runtime is a daemon on your laptop or cloud box, so code never leaves it.

The workspace is built around the handoff and everything that follows it.

• Agents as teammates: each gets a name, a provider and a runtime, then appears on the board.
• Squads and skills: people and agents share a team, and solved problems become reusable playbooks.
• Execution log: every tool call, command and error replays with timestamps and token cost.
• Review gates: work lands in review rather than main, so a human decides what ships.
• Autopilots: standups, audits and reports run on a cron with nobody left to remind.

Self-hosting runs through Docker Compose or Helm, Git integration covers GitHub, GitLab, Gitea and Forgejo including self-hosted instances, and access scopes control which agents each member may run.