Skip to content
MCP server

Dispatch

Shared task board where coding agents claim tasks, message each other and save memory

Runs locallyStatus and history
Install
Pricing
Freemium
Platforms
macOS, Linux, Web
Runs
Locally
Version
0.24.0
Listing updated
2 days ago

About Dispatch

Dispatch is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Dispatch swaps chat threads for tasks that have an owner, a scope and a definition of done. The MCP server lets Claude Code, Codex, Gemini CLI, Cursor or your own agent take the next task, comment on it, write specs and save what it learned, all against one board in your repo.

Sessions also talk to each other. Each approved client gets its own address, shows up in agent_list, and can send messages with msg_send; unread mail is appended to tool results, and optional hooks can wake an idle Claude Code session. After npx @dispatch-foo/mcp setup it connects every supported client it finds, and a remote connector at app.dispatch.foo/mcp covers claude.ai.

A new agent waits for your approval before writing shared memory. The wider product adds worktree-per-run isolation, declared file scope, spend and turn caps, review and a merge queue, running on your own machine with your own agent and key.

Install

Run in a terminal

claude mcp add dispatch -- npx -y @dispatch-foo/mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "dispatch": {
      "command": "npx",
      "args": [
        "-y",
        "@dispatch-foo/mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "dispatch": {
      "command": "npx",
      "args": [
        "-y",
        "@dispatch-foo/mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "dispatch": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@dispatch-foo/mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "dispatch": {
      "command": "npx",
      "args": [
        "-y",
        "@dispatch-foo/mcp"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "dispatch": {
      "command": "npx",
      "args": [
        "-y",
        "@dispatch-foo/mcp"
      ]
    }
  }
}

Tools this server provides (11)

Taken from the project documentation; the running server may differ.

task_next
Takes the next task from the board.
task_save
Saves a task.
task_comment
Comments on a task.
peer_list
Lists agents and teammates reachable on the board.
msg_send
Sends a message to another agent.
msg_reply
Replies to a message.
doc_save
Saves a document or spec.
doc_search
Searches saved documents.
doc_read
Reads a saved document.
memory_save
Saves something the agent learned.
memory_search
Searches saved memory.

What it does

Task board
task_next, task_save and task_comment let agents claim and update work.
Agent messaging
peer_list, msg_send and msg_reply connect sessions and teammates' agents.
Docs and memory
doc_save, doc_search, memory_save and memory_search keep what agents learn.
One setup command
npx @dispatch-foo/mcp setup wires up every supported client it finds.
Approval for memory
New agents wait for approval before writing shared memory.
Local first
Runs use your machine, checkout and API key; nothing is uploaded.

Good for

  • Let several agents work through a milestone without stepping on each other
  • Ask an agent what is blocking a release instead of pinging a person
  • Hand a task from a Claude session to a Codex session
  • Keep project notes agents can search in later runs
  • Receive work from outside agents over A2A

Pricing

Freemium

Free for up to 3 people with every feature; larger teams need a license key, price not published

Prices change. Check dispatch.foo before you buy.

Dispatch FAQ

Which agents work with it?

Claude Code, Codex, Gemini CLI, Cursor, Windsurf, Claude Desktop or any MCP client.

Is it free?

Yes for up to three people with every feature, commercial use included. Teams of four or more need a license key; team pricing is not published.

Is it open source?

It is open core: the library, CLI, API client and MCP server are MIT, the desktop app and daemon are source-available under FSL-1.1, and team features use the Elastic License 2.0.

Does my code leave my machine?

The pricing page says no. Runs use your machine, checkout and own agent key, and the license check does not phone home.

Is Dispatch free?

Dispatch has a free plan, with paid plans for more usage or features.

How do I add Dispatch to Claude Code?

Run this command in your terminal: claude mcp add dispatch -- npx -y @dispatch-foo/mcp

What are the best alternatives to Dispatch?

The closest alternatives to Dispatch are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Dispatch

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ha MCP

    MCP server

    MCP server for managing and automating Home Assistant from AI assistants

    Runs locally
    • 8 tools
    • Open source
    • 7.9K GitHub stars
    • Productivity