Plugins

Bridges the CursorTouch/Windows-MCP desktop automation server into DSH through the official MCP client bridge, registering its screenshot, UI tree snapshot, mouse and keyboard input, clipboard, process and registry tools as mcp__windows__* tools for the model, and adds a windows_mcp_doctor diagnostic tool. Computer Use is off by default and is switched on from the plugin panel.

Lets the DSH agent operate your Windows apps and web pages: it reads controls through UI Automation, clicks, types, captures the screen and reads text with OCR, and works on web pages in its own browser in the background, without moving your mouse; when it needs your mouse or keyboard, an on-screen indicator appears and Esc stops it.

Let DeepSeek Harness see and operate any program on Windows. The model looks at the real screen and drives the mouse, keyboard and windows itself. It is fast: several steps run in a single round trip, and a window can be focused before input lands, so the click hits the right one. It can do anything you would do on a desktop, including but not limited to CAD drawing, circuit simulation, and sending messages.

Accessibility-first macOS computer use: fresh observations, stale-state rejection, scoped permissions, and safe input.

Cross-platform Computer Use for DeepSeek Harness: virtual-mouse human-like operation (screen_observe + computer_click + type + key + scroll + drag, 11 tools), AX-tree zero-vision-cost mode with free GLM-4V-Flash vision fallback, safety guards (snapshot TTL / app whitelist / dangerous-action approval / password-box protection).

Text-first computer use for DSH: background Chromium control via Playwright/CDP plus accessibility-first macOS control; actions stay pinned to the right process and window without taking the user's pointer, ships a Developer ID signed, notarized Universal 2 DMG.

Vision-first Mac computer use, for when the accessibility tree is unavailable. Ships 11 desktop_* tools - window list, screenshots, OCR text targeting, numbered elements, coordinate clicks, typing, scrolling, drag - plus a Codex-style cursor overlay with spring animation. Clicks are delivered to the target process, so the real pointer never moves and nothing steals focus. In DSH it shows up as Mac Desktop Control with a settings page that checks the engine and the Screen Recording / Accessibility permissions, with a button that opens the matching System Settings pane. Works out of the box: the tarball bundles the dsh-cu engine and the overlay as prebuilt universal binaries (Apple silicon + Intel, macOS 11+, ad-hoc signed), compiled from the repo's Swift sources by the release workflow - no clone, no Xcode, no install.sh. The only setup is granting DeepSeek Harness Screen Recording and Accessibility once. macOS only; on Windows and Linux the plugin loads without registering the desktop tools.

AX-first macOS computer use for DeepSeek Harness: accessibility snapshots with stable refs, background input, Unicode typing, menus and apps, set-of-mark screenshots, on-device OCR and verified actions — 11 tools bridged to the macos-cu CLI.

Windows desktop control through UI Automation, window screenshots and mouse/keyboard input, with per-session app consent and action approvals.

Windows Computer Use for DeepSeek Harness with 21 tools and verified targeting: controls are addressed through their UI Automation identity and patterns, and numbered marks are checked against the live window before input, so a stale target is refused instead of clicked. A known sequence runs in one computer_batch call, one model round trip; incremental observations, compact receipts (normally no more than about 6 KiB extra), bounded handoff waits (HANDOFF_WAIT_TIMEOUT/HANDOFF_HELD_TIMEOUT as continuations), the desktop-block cycle boundary and a deterministic 100 s owner-scoped leave backstop keep context small and prevent a stuck handoff from waiting out the tool deadline. Approval modes settle an unanswerable ask honestly: in a danger-full-access session with no approval card the action runs under that grant, every other sandbox — and every delegated subagent session — refuses and names the unavailable channel instead of reporting a user rejection, and gate diagnostics land in policy.jsonl. Human handoff, a model-visible stop brake and prebuilt native workers.

AX-first macOS computer use for DeepSeek Harness: accessibility snapshots with stable refs, background input, Unicode typing, menus, apps, set-of-mark screenshots, on-device OCR, verified actions, and optional Jev (TypeSafe System One) semantic guards — 1

Official DSH plugin for nuphus-mcp (by the nuphus-mcp maintainers): desktop and browser automation (38 tools) with PaddleOCR element perception and Chrome CDP browsing, via a persistent stdio MCP child process.

Frontend GUI debugging and testing for DeepSeekHarness agents: 36 browser_* tools for structured GUI automation, visual comparison, dynamic frame analysis and network debugging on Web/Chromium.

Auditable vision and cross-platform Computer Use runtime for DeepSeek Harness with source-preserving evidence.

Windows desktop observation and control: accessibility tree, screenshots, OCR fallback, mouse and keyboard, with session-scoped approvals and an ESC halt gate. Computer use and desktop automation for DSH, including GUI automation fallbacks.

Windows desktop control exposed as native DSH tools - whole-screen and single-window capture (the window one goes through PrintWindow, so a covered window still captures), click, drag, typing, keyboard shortcuts, scrolling, window management, pixel and window polling, and UIA queries that can invoke a control or read its value instead of clicking coordinates. 13 tools over a PowerShell 7 child process; no MCP server, no resident process, no API key.

One MCP tool surface over two Windows desktop-automation backends (dsh-computer-use-win and windows-mcp). Both must be installed first: dsh-computer-use-win as a DSH plugin, windows-mcp via uv tool install windows-mcp --python 3.14, then run the bundled scripts/patch-backend.mjs once (it rewrites the local windows-mcp source; without it screenshot/read_webpage/launch_app fail) and disable the standalone wincu mount. It merges their 44 tools into 29 by dropping 10 duplicates, 4 dangerous ones (PowerShell, Registry, FileSystem, Process) and 1 tool-management entry, and blocks dangerous calls before forwarding through a reversibility-graded guard.

Windows computer-use for DeepSeek Harness. The computer tool uses ChatGPT Windows Computer Use action names, reusable window targets, UI Automation, occlusion-immune window screenshots, background-first verified input, and an isolated Chromium browser route.

macOS desktop control for DeepSeek Harness: agent operates non-browser apps (CapCut, Photoshop, WPS, native clients) like a human — 14 tools for mouse, keyboard, scrolling, app activation, windows and screenshots, plus a vision closed-loop (desktop_see for image understanding + desktop_locate for UI element coordinates) that reuses dsh existing vision credentials. GUI-session service starts on demand, no login autostart.

Computer-use for DSH on Windows: a session-scoped toggle in the composer that registers nine tools for screenshots, mouse clicks and drags, keyboard typing and shortcuts, window switching, UI Automation element lookup, window-to-account mapping, a batch prompt queue and download collection. Desktop automation input is native Win32 SendInput and defaults to human-like pacing - bezier cursor paths with overshoot correction, real VK keystrokes, log-normal typing with occasional typo corrections and a pause before each click - with realistic, fast and custom (0.2-8x) speed presets. Windows only. Off by default and switched on per session by the user; screenshots are sent to the model, and clicks on send/pay/delete-like controls are refused unless confirmed.

Windows desktop control for DeepSeek Harness: accessibility-tree snapshots, click by control name, delivery verdicts, and repeatable task files.

Config-only DeepSeek Harness bundle that mounts the in-box MCP client bridge against CursorTouch/Windows-MCP, excluding PowerShell, FileSystem, Registry, Process and Scrape by default (16 of the 21 tools upstream 0.8.8 publishes). Requires uv on PATH.

DSH Cordis computer-use provider for the ZTE Nde desktop. npm install auto-fetches the musl binary built by computer-use-linux-zte-nde (v0.7.7-nde.3) into the package's own bin/ via postinstall; resolveCommand() resolves the cordis command path against import.meta.url so the mount works regardless of dsh web's cwd. Reserves the shared computer-use slot via ctx.computerUse.register and exposes MCP tools under mcp__computer-use-linux-nde-mcp__* (screenshot, list_windows, click, type_text, doctor, etc.). Two notable Nde fixes upstream of this provider: KWin caption+coords fallback for window targeting, and SnapshotCache that keeps AT-SPI single-wake per (app, pid) so Chrome is not repeatedly disturbed.

Windows desktop control through UI Automation: nine tools that read a window control tree and drive elements by id through their own UI Automation patterns, with every write gated by an approval policy, an append-only action log and a settings panel.