Summary

AgentRQ is a high-performance platform designed for real-time, human-in-the-loop collaboration with AI agents. It addresses the challenge of integrating AI into existing task management workflows by enabling AI models to directly interact with workspace tasks via the Model Context Protocol (MCP). This allows AI agents to autonomously pull, update, and manage tasks, while facilitating seamless communication and permission requests with human operators, significantly boosting productivity for developers and researchers managing complex AI-driven projects.

How It Works

AgentRQ employs a decoupled service-oriented architecture, featuring a Go/Fiber backend API, an MCP server, and a Vue.js 3 frontend. Its core innovation lies in the Model Context Protocol (MCP), which exposes workspace tools and resources to AI models. Agents "see" the workspace state through MCP, enabling them to autonomously execute tasks, update statuses, and communicate. This approach ensures real-time synchronization and a unified view for both human operators and AI collaborators.

Quick Start & Requirements

Prerequisites include Go 1.21+, Node.js 18+ (npm), and a Google Cloud Console OAuth2 Client ID/Secret. Local setup involves running make install followed by make dev. The frontend is accessible at http://localhost:5173. Docker images are available for self-hosting. Specific setup guides detail integration with Claude Code, Gemini CLI, and OpenAI Codex via MCP configuration files (.mcp.json, .claude/settings.local.json, .codex/config.toml).

Highlighted Details

• Model Context Protocol (MCP): Enables AI models to directly interact with workspace tasks, resources, and tools for autonomous operation.
• AI Agent Gateways: Facilitates integration with various AI agent frameworks (ACP, Codex) for real-time task notifications and command execution.
• Supervisor (CoreMCP): Provides a global MCP server for cross-workspace management, task aggregation, and administrative control across the entire AgentRQ account.
• Official Extensions: Pre-built plugins simplify integration with popular AI agent CLIs like Claude Code and Gemini CLI.
• Slack Integration: Offers real-time task creation, thread replies synchronization, and agent permission request handling.

Maintenance & Community

No specific details regarding maintainers, community channels (like Discord/Slack), or active sponsorships were found in the provided README.

Licensing & Compatibility

The project is licensed under the Apache-2.0 license, which is permissive and generally suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

Initial setup can be complex, requiring detailed configuration for Google OAuth2 and specific AI agent integrations. The platform's effectiveness is dependent on the capabilities and integration points of the connected AI models and agent frameworks.