Plugins

Lightweight local visual grounding engine exposed as agent tools

Wires the local cc-computer-use MCP server into DSH with one process per root session, giving each session its own Xephyr sandbox screen and private AT-SPI bus (requires the cc-computer-use MCP server).

macOS computer use with eight computer_* tools: accessibility-tree and Vision OCR screen reading, actions by element index or visible label that also reach background windows, and CoreGraphics mouse and keyboard events with a resident low-latency daemon for game-grade input sequences; its Swift helpers compile on first use, so there is no runtime dependency.

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

Edge glow for DeepSeek Harness: blue while it drives your mouse and keyboard, amber while it asks a question, red while it waits for approval, plus a top banner naming the current action. Desktop only: the overlay is a separate Electron process (auto-detected from the DSH desktop app or dsh-pet; override with DSH_ELECTRON_PATH) and stays dark without one.

Computer Use for every DSH session: screen capture, mouse and keyboard, app launch, plan and skill replay, an audit log, guards for dangerous actions, a background code agent, and an OCR workflow (text and coordinates) for apps whose accessibility tree is empty. Verified on macOS only: Windows has a basic GUI backend that is unverified, Linux has no desktop backend (calls fail with unsupported_platform), and the window/AX/browser/clipboard tools register on macOS only.

Visibility layer for DSH's built-in Cua Driver computer-control channel: a top banner names the controlling session and its goal, Esc truly stops the agent, and four precision tools click more accurately - they check that the point is not covered by another window, verify what the click changed, find a control by its name, click up to five controls in a row, and wait for a control to show up; a control can also be addressed by element id and run in the background when the application supports it. When the official channel is missing it asks whether to install it and can install it in one click. Windows only.

English port of Windows Computer Use for DeepSeek Harness: 10 desktop tools (computer, computer_batch, open_application, windows, ui_elements, clipboard, switch_display, app_skill, request_access, list_granted_applications) with UI Automation, per-app access control, English settings UI and mini progress card overlay, Esc pause/resume, and DSH 0.2.1 unpacked koffi.node auto-discovery. Fork of copylee711/dsh-computer-use; it reuses the same plugin id and tool names, so the two cannot be installed together.

Computer use for DeepSeek Harness: js and js_reset tools that read windows, click, type, and screenshot macOS apps through the locally installed ChatGPT app. Apple Silicon macOS only; requires the ChatGPT desktop app.

Computer Use enhancement plugin for DeepSeek Harness (DSH). 纯 JS 视觉差分、网格位置编码、长程任务闭环与人机共驾授权。

Windows x64 only, 0.5.4 preview: two scoped tools (dragon_browser and dragon_desktop) that drive a visible browser through the DOM or CDP and Windows apps through UI Automation, with per-origin and per-executable read/control scopes, batch steps, a physical-Esc emergency stop and a redacted action log. 0.5.4 only widens the DSH compatibility range to 0.2.0-rc.2 through 0.2.1.x, because 0.5.3 pinned it to an exact DSH version and was silently skipped on DSH Desktop 0.2.1. Two long-lived native processes remove the per-action process spawn, so a single key step takes about 0.7 s on the same machine; when another window steals the foreground the step is aborted instead of typing into the wrong place. The browser side ships its own Edge/Chrome launcher and CDP client, so it no longer needs another browser plugin or the Electron runtime. It is not an OS sandbox; logins, admin windows and secure desktops stay with the user.

Installs the codex-cua skill so DeepSeek Harness can use Codex computer use: drives Chrome and native macOS apps through the runtime bundled with the ChatGPT/Codex desktop app, ships a persistent session client and a doctor self-check. Requires the ChatGPT/Codex desktop app with its Chrome extension, Python 3.9+, and macOS.

Lets the agent drive Windows by sight: Windows OCR turns on-screen text into clickable targets, then clicks, drags, scrolls, types and presses keys, verifying each action with a per-pixel diff and an OCR text delta; F6+F7 is a separate emergency stop.

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.

macOS cross-application desktop tools for DSH using CuaDriver, with a hidden agent cursor, background window selection and post-action observations.

Bridges the Cua Driver MCP server into DSH: installs the cua-driver executable when it is missing, starts it over stdio through the in-box MCP client, and registers its 60 accessibility-first window, input, menu, and browser tools with a bundled workflow skill and one-time macOS permission guidance.

Give DeepSeek Harness its own persistent computer: a bundled microVM with shell, browser, live view and approval gates

Control Proxmox VE virtual machines from the conversation: 16 tools for the VM console over VNC (screenshots, mouse, keyboard, batched actions, Unicode text and clipboard) and for running commands and reading or writing files through the QEMU guest agent, all over one ssh connection to the node, so the host needs no X11 or Wayland.

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.

Drive the desktop and Chrome from DeepSeek Harness using the computer-use runtime inside the ChatGPT desktop application.

Windows desktop control with window-bound screenshots and accessibility state, text readback, DSH approvals, and activity overlays.

macOS computer-use tools backed by a plugin-owned Cua Driver runtime.

TOFU BOX 伞包：一条命令装齐 tofu-zyc 的全套 dsh 插件（model-search、model-tuning、read-image-preview、shell-selector、computer-use、mcp-ui）。本包不含代码，只做两件事：以 dependencies 拉入六个子包，以 patch 层注册它们的插件行——即 dsh-base 同款 bundle 模式。装本包或单装子包，二选一，不要混装。