Deskbrid v1.0.0 lets AI agents control Linux desktops via MCP
A new open-source tool called Deskbrid gives AI agents the ability to see and interact with a Linux graphical desktop. It includes a built-in MCP server, so tools like Hermes Agent can connect directly and control windows, buttons, and files on screen. For Linux users, this opens a path to delegating GUI tasks to an AI agent without writing custom automation scripts.
Deskbrid acts as a bridge between AI agents and the Linux graphical desktop (GUI). Instead of an AI only being able to read text or call APIs, Deskbrid lets it actually click buttons, open windows, and move files — just like a human sitting at the keyboard. Version 1.0.0 supports 9 compositor backends, meaning it works across the major Linux display systems including both X11 and Wayland.
The built-in MCP server is the key feature for Hermes Agent users: any MCP-compatible AI tool can connect to Deskbrid and issue desktop commands. For example, you could ask Hermes Agent to 'open this folder and rename the files' and the agent would carry it out through the actual GUI. The project is Linux-only and this is its first stable release, so thorough testing is recommended before relying on it for important tasks.
Key points
- Lets AI agents click, type, and navigate a Linux desktop's graphical interface
- Built-in MCP server means Hermes Agent and other MCP-compatible AIs can connect directly
- Supports 9 Linux display systems (including X11 and Wayland) out of the box
- Enables AI-driven GUI automation without writing custom scripts for each app
- Linux-only; first stable release — test carefully before using in critical workflows
Quick term guide
- open-source
- Software whose code is shared publicly so others can inspect, use, or change it.
- AI agents
- AI agents are AI tools that can carry out steps toward a goal, not just answer once.
- built-in MCP server
- A built-in way for AI tools to connect with apps or data sources.
- MCP server
- A server that helps AI tools connect to outside services in a standard way.
- Hermes Agent
- It appears to be a tool or community for building and managing AI agents.
- automation
- A way to make repeated work happen without doing every step by hand.
- compositor backend
- The software engine on Linux that draws windows and graphics on your screen; different Linux setups use different ones.
- Interface
- The visual parts of a program that a human interacts with.