AgentRQ

Prerequisites

• Go 1.21+
• Node.js 18+ (with npm)
• Google Cloud Console: An OAuth2 Client ID and Secret.

1. Install all dependencies

make install

🛠 Getting Started

Installation

npm install -g @agentrq/acp-gateway

Installation

npm install -g @agentrq/codex-gateway@latest

Setup

1. Configure agentrq MCP server for Codex (project-level)

Codex reads project-level MCP server config from .codex/config.toml. Create this file so the Codex agent can use agentrq tools directly during task execution (replace <WORKSPACEID> and <TOKEN> with your values from the agentrq dashboard):

mkdir -p .codex
cat >> .codex/config.toml << 'EOF'

[mcp_servers.agentrq-workspace]
url = "https://<WORKSPACEID>.mcp.agentrq.com/?token=<TOKEN>"

[mcp_servers.agentrq-<ID>.tools.updateTaskStatus]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.getWorkspace]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.reply]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.createTask]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.downloadAttachment]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.getTaskMessages]
approval_mode = "approve"

[mcp_servers.agentrq-<ID>.tools.getNextTask]
approval_mode = "approve"
EOF

2. Configure the gateway's agentrq connection

Create a .mcp.json in your project root so codex-gateway can connect to the same agentrq workspace:

{
  "mcpServers": {
    "agentrq": {
      "type": "http",
      "url": "https://<WORKSPACEID>.mcp.agentrq.com/mcp?token=<TOKEN>"
    }
  }
}

Note: .mcp.json is used by codex-gateway to receive tasks. .codex/config.toml is used by the Codex agent itself to call agentrq tools (e.g. reply, updateTaskStatus) during execution.

Usage

1. Ensure you have a .mcp.json in your project root.
2. Run the gateway followed by your agent's ACP command:

```bash

Usage

Run codex-gateway from your agentrq workspace root (the directory containing .mcp.json):

```bash

Configuration

1. Create a _config/base.yaml (or development.yaml) in the backend directory.
2. Fill in your Google OAuth2 credentials:

auth:
  google:
    client_id: "your-google-client-id"
    client_secret: "your-google-client-secret"

Step 2 — `.claude/settings.local.json`

Add a .claude/settings.local.json file in the same project directory to pre-approve the AgentRQ tools and avoid permission prompts on every action:

{
  "permissions": {
    "allow": [
      "mcp__agentrq-WORKSPACE_ID__updateTaskStatus",
      "mcp__agentrq-WORKSPACE_ID__getWorkspace",
      "mcp__agentrq-WORKSPACE_ID__reply",
      "mcp__agentrq-WORKSPACE_ID__createTask",
      "mcp__agentrq-WORKSPACE_ID__downloadAttachment",
      "mcp__agentrq-WORKSPACE_ID__getTaskMessages",
      "mcp__agentrq-WORKSPACE_ID__getNextTask"
    ]
  },
  "enableAllProjectMcpServers": true,
  "enabledMcpjsonServers": ["agentrq-WORKSPACE_ID"]
}

🤖 Claude Code & AI Integration

AgentRQ is designed for seamless integration as a Claude Channel. This allows your AI agents to see tasks assigned to them and respond directly within your Claude session.

Each workspace has its own MCP URL and token (visible in the workspace setup modal). In production, these follow the pattern https://WORKSPACE_ID.mcp.agentrq.com/.

🧩 Official Extensions

AgentRQ provides official extensions for major AI agent CLI tools to simplify setup and integration with its supervisor MCP. The sub agents MCPs should use their own workspace specific MCP server URLs.

🔌 Integrations

Slack Integration

AgentRQ supports multi-tenant Slack integration for real-time task creation, thread replies sync, and agent permission requests: - Slack Integration Setup & Usage Guide