Human-in-the-Loop Task Manager for AI Agents

Security Guidelines for AgentRQ Workspaces

Every agent connects to its AgentRQ workspace with an MCP URL, and that URL carries the workspace's token. Anyone who holds it can act as your agent in that workspace. These guides cover what to do when that credential leaks, step by step, and how to keep it from leaking again.

Where to Find These Settings

Everything in these guides lives in your workspace: open it, go to Settings and choose the Danger Zone tab.

Start Free