GitHub Repository

Give any AI agent a full desktop — it sees the screen, clicks, types, and runs apps like a human. Automate anything with a UI: browsers, legacy software, internal tools. No API needed. One Docker command.

127 starsPython

GhostDesk – MCP server giving AI agents a full virtual Linux desktop

Name: GhostDesk – MCP server giving AI agents a full virtual Linux desktop
Availability: InStock
Author: maltyxxx

by maltyxxx·Mar 25, 2026·5 points·0 comments

Visit Project View on HN

AI Analysis

●●●BangerWizardryZero to One

Gives AI agents a full Linux desktop with human-like mouse movement to bypass bot detection.

Strengths

•MCP server integration works with Claude, GPT, Gemini, and local models like Ollama
•Human-like input simulation bypasses bot detection that breaks standard Selenium scripts
•Runs in Docker with support for parallel instances driven by sub-agents

Weaknesses

•Linux-only support limits use cases for Windows or macOS specific software
•Relies on LLM vision capabilities which can be slow and expensive at scale

Post Description

Most LLMs can reason. They can't use software.

GhostDesk gives your agent a full Linux desktop and the motor skills to operate it like a human — realistic mouse movement, natural typing, screenshot fallback for CAPTCHAs. It reads UIs semantically and behaves like a real user when sites try to detect bots.

Book a flight, scrape a site without selectors, operate legacy software with no API, run QA across an entire app — one prompt. If a human can do it on a desktop, your agent can too.

Runs in Docker. Spin up multiple instances in parallel, each driven by a sub-agent. No real ceiling.

Works with Claude, GPT, Gemini, or any local model (Ollama, LM Studio). MIT.Most LLMs can reason. They can't use software.