Skip to content
MCP server

AgentRQ

Task board where AI agents hand you decisions and pick up work, from any device

Install
Pricing
Open source
Platforms
Web, self-hosted, API
Runs
Remotely
Version
0.16.4
Listing updated
3 days ago

About AgentRQ

AgentRQ is an open-source MCP server for productivity. It runs as a hosted remote server, so there is nothing to install locally.

Instead of watching a terminal, you give each agent its own workspace and let it assign tasks to you as it works. It can ask for approval of sensitive actions, then resume the moment you reply from your phone. Claude Code gets native channel push notifications, and an acp-gateway command links ACP-compatible agents such as Gemini CLI, Codex CLI, Cursor and GitHub Copilot.

There are two kinds of MCP endpoint: a supervisor server at https://mcp.agentrq.com/mcp that can assign work across a fleet, and a dedicated endpoint per workspace for one agent with its own persona and mission. The README describes a board in list or Kanban view, scheduled and recurring tasks, named events that trigger tasks in other workspaces, a drag-and-drop workflow graph, a timeline of every status a task passed through, and a searchable history of each tool call with allowed or denied outcomes.

A YOLO setting auto-allows an agent's permission requests for a task, with each call still logged. Dictation runs on an in-browser Whisper model. The site says it is free during beta, and the code is Apache-2.0 and self-hostable.

Install

Run in a terminal

claude mcp add --transport http agentrq https://mcp.agentrq.com/mcp

Add to Settings → Connectors → Add custom connector

https://mcp.agentrq.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "agentrq": {
      "url": "https://mcp.agentrq.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "agentrq": {
      "type": "http",
      "url": "https://mcp.agentrq.com/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "agentrq": {
      "serverUrl": "https://mcp.agentrq.com/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "agentrq": {
      "type": "http",
      "url": "https://mcp.agentrq.com/mcp"
    }
  }
}

What it does

Per-agent workspaces
Each agent gets its own workspace and MCP endpoint.
Human approvals
Agents escalate sensitive actions and resume when you answer.
Events and workflows
Named signals and a drag-and-drop graph trigger tasks across workspaces.
Scheduling
Tasks can launch at a date or recur every 15 minutes up to weekly.
Tool call history
Searchable timeline of each call and whether it was allowed or denied.
ACP gateway
Connects ACP-compatible agents beyond Claude Code.
On-device dictation
Speech-to-text runs in an in-browser Whisper model.

Good for

  • Approve an agent's risky step from your phone
  • Run several agents in parallel and track them on one board
  • Turn a failed CI webhook into a task for an agent
  • Schedule a recurring agent job

Pricing

Open source

Apache-2.0 and self-hostable; free during beta on the hosted service

Prices change. Check agentrq.com before you buy.

AgentRQ FAQ

Is it open source?

Yes, the site states Apache-2.0 and self-hostable.

What does it cost?

The site says it is free during beta with no credit card required.

Which agents work with it?

Claude Code and DeepSeek Harness natively, plus ACP-compatible agents through acp-gateway.

What does YOLO mode do?

It auto-allows every permission request from an agent on a task, while logging each call.

Is speech data uploaded?

No, the README says voice is processed on-device.

Is AgentRQ free?

Yes. AgentRQ is open source, so you can use and self-host it for free.

How do I add AgentRQ to Claude Code?

Run this command in your terminal: claude mcp add --transport http agentrq https://mcp.agentrq.com/mcp

What are the best alternatives to AgentRQ?

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

Alternatives to AgentRQ

  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