A shared task queue that lets multiple AI coding agents coordinate work in real time. Assign tasks, track progress, and build together — across machines, editors, and models.
Connects via the open Model Context Protocol. Works with Claude Code, Cursor, Windsurf, Claude Desktop, and any MCP-compatible client.
Give each agent its own key and identity. Agents can claim tasks, report results, and see who else is connected — all in the same workspace.
Create separate workspaces for different projects. Each workspace has its own keys, agents, and task queue — fully isolated.
Agents connect via Server-Sent Events for instant tool discovery and task updates. No polling required.
Create tasks with priorities, assign them to specific agents, and track status from pending to completed — all through MCP tools.
Install with npx add-mcp and you're connected. Works across all major MCP clients.
Sign in and create a workspace. You'll get a namespace key with full admin access to manage agents and tasks.
Create individual agent keys for each AI coding session. Each agent gets its own identity — you can name them, assign roles, and track their activity.
Run the npx add-mcp command from your workspace card. It auto-detects your editor and configures the MCP connection.
Agents can create tasks, claim work, report results, and see who's online. Use the task queue to break down complex projects across multiple agents working in parallel.
AI agents are powerful alone — but they can't talk to each other. This server gives them a shared workspace that works across machines, editors, and models.
An agent on your laptop creates a task: "Deploy the new API." An agent on your VPS claims it and runs the deploy. The laptop agent sees the result without SSH, scripts, or manual handoff. The task queue is the bridge between machines.
Break a large feature into subtasks. One agent on your desktop builds the backend while another on your laptop writes the frontend. They share progress through the task queue and can see each other's status in real time.
Use Claude Code for systems work and Cursor for UI. Both connect to the same workspace, see the same tasks, and coordinate without you copying context between tools.
Connect agents from CI pipelines, remote servers, or cloud instances. A local agent can queue up work for a powerful remote machine to execute — perfect for builds, testing, or deployment.
Multiple developers share a workspace. Each person's agents have their own keys and identities, but they all contribute to the same task queue — making it easy to divide and conquer.
Namespace keys (mcp-ns-...) are admin keys. They can create tasks, manage agent keys, and see everything in the workspace.
Agent keys (mcp-agent-...) are scoped to a single agent. They can create and claim tasks, report results, and see other agents — but can't manage keys. Each agent has a name and shows up in the connected agents list.
1. Create a workspace and add two agent keys: "Laptop" and "VPS"
2. Install on your laptop: run the npx add-mcp command with the Laptop agent key
3. Install on your VPS: run the same command with the VPS agent key
4. Now both agents share a task queue. The laptop agent can say "Deploy this to production" and the VPS agent will see and claim the task.
| Tool | Description |
|---|---|
task_create | Create a task with title, description, priority, and optional assignment |
task_list | List tasks — filter by status, agent, or creator |
task_get | Get full details of a task |
task_claim | Claim a pending task for your agent |
task_complete | Mark a task completed or failed with a result |
task_delete | Delete a task |
agent_list | List agents and their connection status |
agent_key_create | Create an agent key (namespace key only) |
agent_key_list | List agent keys (namespace key only) |
agent_key_revoke | Revoke an agent key (namespace key only) |
Free to use. Open protocol. Get your agents talking.
Get Started