desktop-mcp

desktop-mcp

Windows desktop-control MCP server: screenshot, window mgmt, mouse/keyboard input, recording.

Category
访问服务器

README

desktop-mcp

PyPI MCP Registry License: MIT tests

Windows desktop-control MCP server: screenshot, window management, mouse/keyboard input, and ffmpeg screen-recording, config-gated by tool group (input off by default) -- built to the same standard as mcp-factory and rag-mcp: own pyproject, fastmcp server, honest README, real test suite.

Quickstart

pip install desktop-mcp
// ~/.claude.json (or any MCP-host stdio client config)
{
  "mcpServers": {
    "desktop-mcp": {
      "command": "desktop-mcp"
      // or: "command": "python", "args": ["-m", "desktop_mcp"]
    }
  }
}

window/input/record tool groups are env-gated -- input stays off unless you explicitly opt in. See Env vars below for the full table.

Tool groups

Group Tools Default state
observe screenshot, list_windows, get_active_window, window_info always on
window focus_window, move_resize_window, minimize_window, restore_window env-gated, off unless DESKTOP_MCP_ENABLE_WINDOW=1
input mouse_move, mouse_click, mouse_drag, mouse_scroll, key_press, hotkey, type_text env-gated, OFF by default -- requires DESKTOP_MCP_ENABLE_INPUT=1
record record_start, record_status, record_stop env-gated, off unless DESKTOP_MCP_ENABLE_RECORD=1

A disabled group returns a structured policy_refusal error (never a silent no-op, never a crash). Input actions are additionally rate-capped (default 60 actions/min, tunable via DESKTOP_MCP_RATE_LIMIT_PER_MIN) -- exceeding the cap returns a structured rate_limited error.

This is defense-in-depth: harness-level permission prompts are the first gate, but the server itself refuses input/window/record actions unless its own env explicitly enables them, so a misconfigured or overly-permissive harness can't turn on capabilities the operator didn't opt into for this process.

Security

Read this before you register desktop-mcp. This is a desktop-control server; what it can do to your machine is spelled out here plainly so you can make an informed decision. Honesty about the blast radius is the point.

  • The observe group (screenshot, window enumeration) is ALWAYS ON and cannot be disabled. Any caller that can reach this server can capture the contents of any visible window on the machine -- including a password manager, an authenticator/2FA app, a seed phrase, a banking tab, or a private message. screenshot returns a file path to a PNG written to the scratch dir (default %TEMP%\desktop-mcp-scratch), not inline pixels, so captured screen contents also persist on disk until that directory is cleared. Treat the scratch dir as sensitive and run this server only on a machine and under a harness you trust.
  • DESKTOP_MCP_ENABLE_INPUT=1 grants keystroke and mouse injection into the focused window -- this is RCE-equivalent. With the input group enabled, a caller can type commands, press hotkeys, and drive the mouse in whatever window currently has focus (a terminal, a browser, an admin tool). It is OFF by default for exactly this reason; turn it on only when you understand that you are handing the caller the keyboard. When disabled every input tool returns a structured policy_refusal (never a silent action).
  • Least privilege: leave input, window, and record groups off unless a specific task needs them, keep this server on a trusted host, and do not run it elevated (admin) unless you have accepted that an elevated input group can drive elevated windows.

The env-gating, default-off input group, structured refusals, and rate cap described above are the enforced controls; this section states, without softening, what remains reachable by design.

Honest-capabilities table

Every claim below maps to the file that implements it and the test(s) that verify it -- no capability is asserted without a corresponding implementation + test.

Claim Implementation Verified by
Capture a screenshot (monitor / region / window) as PNG desktop_mcp/groups/observe.py::screenshot tests/test_observe.py::TestScreenshot, live: tests/test_live_smoke.py::test_live_screenshot_real_png
Enumerate windows / get active window / look up by title desktop_mcp/groups/observe.py tests/test_observe.py::TestListWindows, TestGetActiveWindow, TestWindowInfo
Focus / move+resize / minimize / restore a window desktop_mcp/groups/window.py tests/test_window.py
Mouse move/click/drag/scroll, key press/hotkey, type text desktop_mcp/groups/input_tools.py tests/test_input.py (mocked pyautogui only -- see Limitations)
Screen recording via ffmpeg gdigrab, hard duration cap, graceful stop desktop_mcp/groups/record.py tests/test_record.py, live: tests/test_live_smoke.py::test_live_record_real_mp4
Input group OFF by default, structured refusal when disabled desktop_mcp/config.py::group_enabled, gated tests/test_config.py::TestGroupEnabled, tests/test_input.py::TestGateDisabledByDefault
Rate cap on input actions (default 60/min) desktop_mcp/config.py::TokenBucket, RateLimiterRegistry tests/test_config.py::TestTokenBucket, tests/test_input.py::TestRateLimit
DPI-awareness bootstrap (per-monitor-v2) so mss/pyautogui coords agree on scaled displays desktop_mcp/config.py::ensure_dpi_awareness tests/test_config.py::TestDpiAwareness (idempotency only; visual coord-agreement is not automated -- see Limitations)
Coordinate validation against virtual-desktop bounds before any mouse action desktop_mcp/groups/input_tools.py::_validate_point tests/test_input.py (out-of-bounds cases)
Orphan-guard: a stale recorder from a crashed process gets killed before a new one starts desktop_mcp/groups/record.py::_orphan_guard tests/test_record.py::TestRecordStart::test_orphan_guard_kills_stale_recorder

Limitations (read before relying on this)

  • UIPI (User Interface Privilege Isolation) -- protects window handles, not input. A medium-integrity process (this server, unless you elevate it) cannot manipulate a window owned by a higher-integrity (elevated/admin) process: focus_window, move_resize_window, minimize_window, and restore_window go through pygetwindow's window-handle APIs (desktop_mcp/groups/window.py), which Windows blocks under UIPI and which this server surfaces as a structured window_action_failed error naming UIPI, never a silent no-op. Keyboard/mouse input tools ARE covered by UIPI too -- and worse, they fail silently. type_text, hotkey, key_press, and the mouse actions (desktop_mcp/groups/input_tools.py) go through pyautogui, whose Windows backend actually calls the legacy keybd_event/mouse_event Win32 functions (not SendInput -- pyautogui tried SendInput and reverted to the older calls; see _pyautogui_win.py in the installed package). Per Microsoft's own docs, UIPI applies to synthesized input generally: "[applications] are permitted to inject input only into applications that are at an equal or lesser integrity level" (MSDN SendInput reference), and critically, "neither GetLastError nor the return value will indicate the failure was caused by UIPI blocking." keybd_event/mouse_event are void-return legacy calls with no failure signal at all, so pyautogui (and this server) cannot detect a block even in principle: if an elevated window has focus, these tools report {"ok": true} while Windows silently discards the injected keystrokes/clicks -- no exception, no error field, nothing reaches the target. This is worse than the window-handle path above, which at least raises a structured window_action_failed error naming UIPI. There is no code-level bypass and no workaround short of running the server elevated (or not enabling the input group at all), which this project does not do or recommend. Operators should treat any elevated app that could plausibly have focus as silently unreachable by this server's input tools, not as a gap that lets them through. Possible future enhancement (not implemented): detect the foreground window's integrity level before injecting and return a structured refusal instead of a false {"ok": true} -- tracked as a v2 idea, not a current capability.
  • DPI scaling. The server sets per-monitor-v2 DPI awareness at startup so mss pixel coordinates and pyautogui point coordinates should agree on scaled displays. This bootstrap is unit-tested for idempotency/no-crash only -- actual coordinate agreement on a live multi-DPI multi-monitor setup has not been automated-tested and should be spot-checked if you're targeting a non-100%-scaled monitor.
  • UAC secure desktop. When Windows switches to the secure desktop (UAC elevation prompts, Ctrl+Alt+Del, lock screen), no process running on the regular desktop -- including this server -- can see or interact with it. Screenshots will show whatever was on the regular desktop before the switch; input calls will not reach the secure desktop at all.
  • Single machine, local only. No network transport, no remote control. stdio only, spawned by the MCP host on the same machine.
  • Input group is off by default in this repo's own registration. See ~/.claude.json's desktop-mcp entry -- DESKTOP_MCP_ENABLE_INPUT is not set there. Enabling it is a deliberate per-registration operator choice, not a code change.
  • pyautogui failsafe. FAILSAFE=True is intentional: slamming the cursor into a screen corner mid-action raises inside pyautogui and aborts the call. This can interrupt an in-flight mouse_drag. Treated as acceptable v1 behavior (see plan's Open questions) -- it's a deliberate human kill-switch, not a bug.
  • No OCR / vision analysis. Screenshots are raw PNGs; interpreting their content is the consumer's job, not this server's.
  • No clipboard tools. Credential-adjacent surface, deferred to a v2 with its own safety design.
  • Not registered with the mcp-factory hub. This ships as a standalone repo (own pyproject, own venv-free system-Python312 install), matching the rag-mcp model. Hub/registry integration is a v2 candidate.

Env vars

Var Effect Default
DESKTOP_MCP_ENABLE_WINDOW enable the window tool group unset (off)
DESKTOP_MCP_ENABLE_INPUT enable the input tool group unset (off)
DESKTOP_MCP_ENABLE_RECORD enable the record tool group unset (off)
DESKTOP_MCP_RATE_LIMIT_PER_MIN input-group rate cap 60
DESKTOP_MCP_SCRATCH_DIR where screenshots/recordings/pidfiles are written %TEMP%\desktop-mcp-scratch
DESKTOP_MCP_LIVE 1 to run real-hardware smoke tests (see Testing) unset (skip)

Usage examples

// A tool call from the MCP host, illustrative -- not a shell command.
{"tool": "screenshot", "arguments": {"monitor": 0}}
// -> {"ok": true, "path": "C:\\Users\\...\\Temp\\desktop-mcp-scratch\\screenshot-....png", "w": 3840, "h": 1080, "monitor": 0}

{"tool": "record_start", "arguments": {"fps": 30, "max_duration_s": 30}}
// -> {"ok": true, "path": "...\\recording-....mp4", "pid": 12345, "fps": 30, "max_duration_s": 30}
{"tool": "record_stop", "arguments": {}}
// -> {"ok": true, "path": "...\\recording-....mp4", "bytes": 800560, "duration_s": 3.13}

// input group disabled (default):
{"tool": "mouse_click", "arguments": {"x": 500, "y": 500}}
// -> {"ok": false, "error": {"type": "policy_refusal", "group": "input", "required_env": "DESKTOP_MCP_ENABLE_INPUT", ...}}

Testing

# unit suite (mocked backends, no real screen/input/recording touched)
python -m pytest -q

# handshake check -- prints every registered tool name
python scripts/list_tools.py

# real-hardware smokes (real screenshot PNG, real ~3s screen recording;
# never input-injection -- see safety rails above)
DESKTOP_MCP_LIVE=1 python -m pytest -q -k live_screenshot
DESKTOP_MCP_LIVE=1 python -m pytest -q -k live_record

Install

pip install -r requirements.txt   # or: pip install .
# deps: fastmcp==3.4.2, mss==10.2.0, pyautogui==0.9.54, PyGetWindow==0.0.9, pywin32==312
# also requires ffmpeg + ffprobe on PATH for the record group

Registered in ~/.claude.json as desktop-mcp (stdio, system Python312, observe+window+record groups enabled, input group absent from env).

Commercial support

Maintained by Jaimen Bell. For production MCP integrations, custom servers, or agent-reliability work, see jaimenbell.dev or sponsor ongoing maintenance via GitHub Sponsors.

<!-- MCP registry ownership marker --> mcp-name: io.github.jaimenbell/desktop-mcp

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选