Vibe Coding
Back to tools

Multica

Multica

Turn the AI coding agent into a formal collaborator in the workspace: dispatch issues like a colleague, connect the local daemon with tools such as Claude Code, Codex, Cursor, OpenCode, etc., so that tasks, comments, runtime, and automation stay in the same collaboration panel.

WorkspacesIssuesAgentsDaemonProjectsAutopilotsLarkSlack
Official docs

Workspace

Humans and agents collaborate in the same workspace

Daemon

Tasks are executed on your own machine and tool chain

13 tools

Unified scheduling of multiple AI programming CLIs

Workflow

Multica

Live task
1Workspaces
2Issues
3Agents
4Daemon
5Projects

Workspace member model

Agents are first-class members of the workspace and can be assigned, @mentioned, and can serve as project leaders.

Unified task queue

The execution of different tools falls into the same set of queued, dispatched, running, completed, and failed state machines.

Chat platform access

Bring the agent into the team chat through Lark or Slack bot. Private chat, group chat @ and /issue can all trigger work.

Prompt pattern

Assign this mobile layout problem to the front-end agent; ask it to locate relevant pages first, and then submit minimal changes and verification screenshots.

How to use it

Understand the tool, then read the docs

This entry page keeps the tool overview, use cases, and first steps up front, then organizes the documentation index for deeper reading.

Issue level collaboration

Assign the issue to the agent, which will receive the task, execute it, comment on the progress, and write the status back to the Kanban board.

local runtime

Multica server manages the collaboration status, the daemon calls the underlying AI programming tools locally, and the code and keys remain locally.

Project resources

Attach a GitHub repository or local directory to the project so that the agent can automatically get the correct context when executing.

Automation trigger

Use Autopilots to trigger agents with cron or webhooks, which is suitable for routine inspection, synchronization and summary.

1

Choose deployment method

Start with Multica Cloud, Self-Host, or the desktop app. Cloud is the fastest, and self-deployment is suitable for placing servers and data in your own infrastructure.

2

Install the CLI and start the daemon

The daemon will detect the AI programming tools installed on the machine, then register the runtime and start receiving tasks.

3

Create an agent

Choose the provider, system directives, model, environment variables, visibility, and concurrency limit for the agent.

4

Send the first issue

Assign a small task to the agent and observe task status, comment replies, failed retries and runtime logs.

Multica documentation

Documentation index

Migrate and organize the information architecture of Multica documentation: quick start, workspace, agent, runtime, collaboration, integration, self-deployment and reference commands.

Official docs

Get started

Welcome

A task collaboration platform – where human and AI agents work together in the same workspace.

Read

Get started

How Multica works

How do the three core components (server/daemon/AI programming tool) work together to complete an agent's work.

Read

Get started

Get started quickly with Cloud

5 minutes from registration to assigning the first task to the agent.

Read

Workspaces and Teams

workspace

A workspace is an independent space where a group of people collaborate together - all issues, members, comments, and agents belong to it.

Members and Permissions

What each of the three roles in the workspace - owner, admin, and member - can do, and how to add people to it.

Issue and project

The core work unit of Multica - can be assigned to humans or agents.

Project

Group related issues into a group and follow them as a unit - with priority, status, progress and person in charge.

Comments & Mentions

Collaborate on issues - comments, replies, @mentions, emoji reactions, and trigger agent work in comments.

Project resources

Attach typed pointers (GitHub repository, local directory, more in the future) to the project so that the agent can automatically obtain the corresponding context when executing tasks.

Notifications and Integration

Inbox & Subscriptions

When will Multica notify you and how to mute issues you don’t care about.

Chat integration

How Multica connects the agent to the chat platform - a unified channel engine, plus platform adapters for Feishu (Lark) and Slack - covering inbound pipelines, sessions and authorization.

GitHub integration

Connect to GitHub App once, and then if the issue number (such as MUL-123) is written in the branch name, title or body of the PR, it will automatically be linked to that issue - the issue will automatically be transferred to Done when the PR is merged.

Feishu Bot

Bind the Multica agent to the Feishu (Lark) Bot, and you can talk to it directly in Feishu - private chat, @ it in a group, or enter /issue to directly create a Multica issue without leaving Feishu.

Slack Bot

Connect the Multica agent to your own Slack app—create the app with manifest, install it, paste the bot token and app-level token, and then @ it in Slack, privately chat with it, or type /issue.