screenshot 29

Eyes for text-only DeepSeek Harness agents: built-in free vision chain + pixel-level tools. Remote vision providers, including the default free fallback, receive image content unless strict local-only use is configured.

Codex-style window capture for DSH Desktop: press both Command/Ctrl keys (or the camera button) to grab the frontmost window on macOS or Windows, attach it to the current chat, and inject available window text as hidden context.

On-demand vision for text-only DSH sessions: images become markers, and a vision_describe tool sends only image + question to an OpenAI-compatible vision model

dsh-plugin-appshot — a DeepSeek Harness plugin for macOS and Windows context screenshot capture.

Screenshot feedback for DeepSeek Harness — let your coding agent SEE what it builds

DeepSeek Harness Computer Use 插件（Windows）：让 AI 看屏幕、点鼠标、敲键盘操控电脑；控制时屏幕泛起橙色光晕并显示“DeepSeek Harness 正在操控你的电脑”，按 Esc 暂停 / 继续。

DSH bundle plugin: 内容创作与多平台分发 —— 桌面截屏/录屏、Markdown→小红书图文卡片（5 配色 × 4 版式、照片插图、背景图）、dev.to 官方 API 发布、发布前人工审阅（草稿落盘），并挂载 Playwright 与小红书 MCP。

DeepSeek Harness plugin that turns UI screenshots into implementation-grade web specs using OCR, deterministic geometry, scene graphs, assets, and render comparison.

Codex-style computer use for DeepSeek Harness (DSH): read the screen and drive the mouse & keyboard. Pairs with picturereader (image_scan/image_ocr) to close the look-act-verify loop. Windows.

Screenshot-to-input for DeepSeek Harness: composer camera button + global hotkey (Alt+A) + listener bound to the app lifecycle, configurable in settings. 截图自动粘贴到 DSH 输入框：相机按钮 + 全局快捷键 + 生命周期绑定 + 设置页配置。

聊天输入框图片工具(自研):上传图片(≤10MB 防烧 token) + 自定义区域截图(Electron desktopCapturer),注入 DSH 草稿图片轨。零外部依赖。

Windows computer-use capability for DeepSeek Harness: screenshot → vision model → simulated mouse/keyboard input, with self-evolving knowledge base.

Point-and-shoot screenshot capture for DeepSeek Harness: clipboard watcher + system floating window (comment & key-point, copy/save-doc/save-image) + zero-config OCR through the host's own multimodal model (Tongyi Qianwen fallback) + Obsidian per-day merg

Unofficial fork of computer-user by jing-hy: Codex-style Windows computer use for DeepSeek Harness. Clicks target live UI Automation element refs (ref/name) rather than pixels estimated off a downscaled screenshot, so they land exactly; screenshots ride t

Built-in browser control for dsh (DeepSeek Harness): lets the AI agent open pages, read content, click, type, scroll and take screenshots in a real local browser (Edge/Chrome via the Chrome DevTools Protocol). Includes a docked live browser pane (CDP scre

PC-Pilot - lightweight Windows + Chromium computer-use runtime, CLI, and DeepSeek Harness host plugin.

Select a range of a DeepSeek Harness conversation and share it as a branded PNG long image, with draggable magnetically snapping range markers.

DSH GUI Toolkit — Windows desktop control tools for DeepSeek Harness (DSH): screen layout, annotated screenshot, single-window capture, click / drag / type / key / scroll, semantic verification, window management, and UI Automation structured queries. Thi

Eyes for your DeepSeek Harness agent — it opens, sees, and fixes what it builds. Headless-browser preview & self-verification tools (screenshot, console, DOM read, interact) as a dsh plugin.

Lets the agent put a picture into the conversation: a screenshot, the clipboard, or a file — every client renders it and the user can open it full screen.

DSH teaching surface: project agent-generated self-contained HTML into a sandboxed in-session board with handwriting annotation and a built-in screenshot that returns to the conversation as a real user message. Ships its own usage skill (skills/stage-pane

DSH 桌面版输入框工具条插件：在「听写」左边加「录屏」与「截图」两个按钮 —— 录屏前先弹「选择要录制的内容」框（屏幕/窗口/自定义区域），截图带选区 + 标注（拖完框选手一松就能画）。Screen capture & recording buttons for the DSH desktop composer.

通用工具图像卡片：read_image 走宿主文件预览；任何结果携带 image 附件的工具（如 computer-use 截图）自动渲染同款缩略图。常驻缩略图、像素尺寸与格式、Ctrl+滚轮缩放与拖动平移、复制路径。接入新图像工具只需加一行 key。

Agent-neutral Windows and Chromium computer-use runtime for Codex, Claude Code, Qoder, Pi, and DeepSeek Harness.

Conversation-style UI/UX visual review plugin for DeepSeek Harness: vision_review / vision_ask tools backed by any OpenAI-compatible multimodal endpoint (e.g. agnes-2.5-flash)

computer-use v2.4：基于 Cua Driver 进程内 SDK 的跨平台桌面操作。13 个工具（截图/窗口列表/缩放/光标/移动/点击/拖拽/滚动/输入/按键/打开应用/连续动作/等待）。v2.4.1 适配 dsh 0.2.0：审批探测改走公有 overrideOf/session 覆盖 + 部署默认策略，任何探测形状失效时保守放行，杜绝 danger-full-access 会话因私有 API 变动而自锁全部工具。v2.4 修「宿主闲置后窗口/应用枚举静默变空」：全部驱动调用统一挂命名

DeepSeek Harness plugin: analyse images out of band with a vision model — pasted images are digested into text before admission and a describe_image tool covers image paths, all without ever changing the session's model.

Display tab for the DeepSeek Harness Web UI: watch and drive the AI's own per-session display. The browser half only talks to the same-origin /api/dsh-display-panel/* proxy (the access token never reaches the browser), and the host half starts the viewer

DSH plugin: browser automation whose click and submit actions are gated by a DSH approval, with the current screen captured as a screenshot the user reviews before the action runs.