AgentRQ, an open-source agent-human collaboration platform using Model Context Protocol, launches with real-time task management and Claude integration.

AgentRQ is a platform that allows AI agents (like Claude) to interact with human operators' task management systems via the Model Context Protocol (MCP), with agents able to see workspace state, receive task notifications, and reply to humans in real-time.

The backend uses Go and Fiber for a REST API, integrates an mcp-go SSE server to expose tools to AI models, and includes a Supervisor (CoreMCP) global MCP server for managing multiple workspaces and tasks across the platform; the frontend is built with Vue.js 3, Pinia, and Tailwind CSS.

Agents connected via Claude Code can use tools such as createTask, updateTaskStatus, reply, getWorkspace, getTaskMessages, getNextTask, and downloadAttachment, and the Supervisor provides listWorkspaces, createWorkspace, listAllTasks, and other management capabilities across all workspaces.

The project is open-source under the Apache-2.0 license and requires Go 1.21+, Node.js 18+, and Google OAuth2 credentials; the frontend runs locally at http://localhost:5173 via a provided Makefile.