"Ghost" matching MCP servers:

Verified control of the whole desktop for agents on Windows and Linux: apps with no API, windows, shell, browser. MCP server + CLI + REST API.

An MCP server that lets agents observe and control an Omarchy or Hyprland desktop: list monitors, windows, workspaces, and accessibility controls, capture screenshots, then move the pointer, type text, send shortcuts, and manage windows and workspaces. It works over stdio with any MCP client, guarding actions with session-lock and single-driver checks while treating on-screen content as untrusted data.

Give any LLM agent a real Android or iPhone. 62 MCP tools: tap, swipe, type, screenshot, screen-tree reading, app launch, camera, TTS, crash reports, batched execution. Android via ADB, iPhone via WebDriverAgent, on-device inference, Docker+KVM emulators. Works with Claude Code, Cursor, LangChain, LlamaIndex, and any MCP client. MIT.

A virtual Linux desktop as an MCP server, shipped in Docker. Agents drive screen, mouse, keyboard across any GUI — browsers, IDEs, office suites, Wine/Windows apps, legacy software — many in parallel.

Enables AI agents to execute shell commands on a host macOS computer with configurable file access permissions, granting access to credentials and tools while enforcing sandbox restrictions.

Enables MCP clients such as ChatGPT and Claude to administer Linux or macOS hosts through authenticated Streamable HTTP, including running shell commands, reading and writing files within a workspace, listing directories, and retrieving system metrics.

Enables control of Ops Desk (Ghost VPS) through 10 MCP tools for alerts, waiting tasks, cron jobs, vault search, and more.

Enables AI-controlled administration of remote Linux hosts and containers, including bounded shell execution, persistent PTY sessions, durable jobs/tasks, hardened filesystem access, process/service control, and resumable transfers.

Enables AI assistants to execute terminal commands on a host machine with configurable, granular permission controls and safety protections. It features multiple security modes, including allowlists and manual approval, to ensure safe command execution within specified directories.

Enables Claude Desktop to control a local host machine through shell, file, git, network, Docker, package, and system monitoring tools, with approval-gated destructive operations.