Multica - An open-source AI Agent team collaboration platform - AiBoss

What is Multica?

Multica is an open-source AI agent team collaboration platform founded by Jiayuan, a former TikTok engineer. The platform transforms coding agents like Claude Code, Codex, and Kimi into true teammates, allowing them to work alongside humans on a Kanban board. These agents can be assigned tasks, autonomously execute code, report progress in real time, and accumulate reusable skills. Multica uses a Linear-like Kanban management interface, supports multiple workspaces and local/cloud runtimes, is vendor-neutral, and self-hostable. It addresses the pain point of "human-to-human AI collaboration being impossible," enabling hybrid human-machine teams to operate as efficiently as traditional teams.

Multica's main functions

Agent as a teammateJust like assigning tasks to colleagues, issues can be assigned to agents such as Claude Code, Codex, and Kimi. Each agent has its own profile and is displayed on the dashboard alongside humans.
Autonomous executionComplete task lifecycle management (enqueue, claim, start, completion/failure), with real-time progress push via WebSocket.
Reusable SkillsEach solution is automatically saved as a reusable skill for the team, supporting scenarios such as deployment, migration, and code review.
Unified RuntimeA dashboard manages the local daemon and cloud runtime, and automatically detects the installed agent CLI.
Multiple workspacesTeams are isolated workspaces, each with its own independent agent, issue, and configuration.

How to use Multica

Install CLI:run brew install multica-ai/tap/multica Or the installation script for the corresponding system.
One-click initialization:implement multica setupComplete the configuration, browser authentication, and start the local daemon.
Verify RuntimeIn the Web client, go to Settings → Runtimes to confirm that your local machine is online as an available Runtime.
Create AgentGo to Settings → Agents → New Agent, select Runtime and Provider (such as Claude Code, Kimi, etc.) and name them.
Assign tasksCreate a new issue on the Board and assign it to the agent. The agent will automatically retrieve, execute, update the status, and submit the results.

Multica's core advantages

Manufacturer neutralityIt does not bind to any single model or agent and supports Claude Code, Codex, GitHub Copilot CLI, OpenClaw, OpenCode, Hermes, Gemini, Pi, Cursor Agent, Kimi, Kiro CLI, etc.
Open source self-hostingThe code and deployment are completely controllable, avoiding vendor lock-in.
Humans and machines have equal rightsThe agent is not an assistant in the chat window, but a formal member of the board, who can post comments, report blockages, and change status.
Skills CompoundingTeam capabilities increase with usage, rather than starting from scratch each time.
Low-frequency intervention designA collaborative language based on issue/kanban/state flow, which is more suitable for large-scale agent parallelism and avoids context explosion in IM-style interactions.

Multica's project address

Project official websitehttps://multica.ai/
GitHub repositoryhttps://github.com/multica-ai/multica

Multica's Competitive Product Comparison

Dimension	Multica	Vibe Kanban	Paperclip
core scenarios	Multi-person team collaboration with AI agent	Single-user local multi-agent parallel	Unmanned company/pure agent autonomous
User Model	Multi-user teams + role permissions	Single developer	Single Board Operator
Interaction methods	Issue + Chat Dialogue	Issue + Local Board	Issue + Heartbeat
Deployment method	Cloud-first + self-hosted	Local-first	Local-first
Management Depth	Lightweight (projects/tags/skills)	Lightweight Kanban	Re-evaluate governance (organizational structure/budget/approval)
Ecosystem binding	Vendor neutral, multi-agent support	Local agent shell	Proprietary plugin system

Multica application scenarios

Multi-Agent Parallel DevelopmentThe development team can incorporate various coding agents such as Claude Code, Codex, and Kimi into the daily project management and version iteration process to achieve parallel human-machine development.
Agent capacity optimizationThe AI Native team can monitor the task status and idle time of each agent through a dashboard, systematically reducing the agent idle rate to improve overall productivity.
Human-machine hybrid collaborationHuman-AI hybrid teams can allocate, track, and review task progress and deliverables from human members and AI agents within a unified collaborative space.
Cloud-based maintenance-free hostingRemote or distributed teams can directly use Multica cloud runtime to host agent execution environments, eliminating the cost of local machine configuration and maintenance.
Team Skills AccumulationThe technical team can turn high-frequency operations such as code review, deployment and data migration into reusable skills, enabling the organization's capabilities to grow exponentially.