84 results
面向金融与产业研究的一屏式实时行情大屏：A股/港股/美股指数、大宗商品、美债收益率、板块热点、主力资金流、7×24 快讯、产业链自选股、AI 大模型 Token 追踪。A real-time market research cockpit on a single screen: CN/HK/US indices, commodities, treasury yields, sector hotspots, capital flows, 7×24 news, industry-chain watchlists and AI token usage trends.
AI 招聘副驾 —— DeepSeek Harness profile bundle：注册招聘工作流 skills（岗位梳理、双通道寻源初筛、约面试、简历评估、台账与日报），并在 Web UI 右侧提供一只 boss/liepin 浏览器观察面板（默认折叠只读，日常动作走 CLI）。
Accessibility-first macOS Computer Use capability for DeepSeek Harness with stale-observation protection, app leases, confirmations, screenshots, and Web diagnostics
Install and launch the dsh-desktop-windowos desktop shell: auto-downloads and auto-updates the exe from GitHub Releases, creates desktop shortcuts (app + web UI), and exposes a desktop_launch tool. First run of the unsigned exe shows Windows SmartScreen - click More info > Run anyway.
Windows computer use for DeepSeek Harness: an MCP stdio server + PowerShell UIA backend, bridged into DSH via @deepseek-ai/dsh-mcp-client. Read/act on real Windows desktop apps (UI Automation tree, screenshots, typed input, OCR, window management).
Mobile adaptation for the DeepSeek Harness Web GUI: single-column layout, sidebar drawer, compact header/composer, fullscreen settings, trajectory floating details, stats pill panel
Mobile layout fixes for the DeepSeek Harness Web UI: settings panel and sidebar on narrow (≤700px) screens. Pure client-side CSS overlay.
Mobile-adaptive DSH web UI: on narrow screens the sidebar rail is hidden and the directory opens as an overlay drawer, so the conversation gets the full width.
A floating whale desktop pet for the DeepSeek Harness Web UI: 95 categorized animations, event-aware actions, custom uploads, click reactions, dragging and screen wandering.
Minecraft Launcher built on DeepSeek Harness: full-screen launcher UI (root slot) with Microsoft device-code login, version download/install, and game launch from the DSH host process.
Focus Mode for the dsh web GUI: a full-screen reading overlay that hides chrome and folds the AI tool-call flow into summaries, reusing the official Markdown / image primitives.
A floating desktop pet for the DeepSeek Harness Web UI: idle breathing, occasional direction turns, random actions, and screen wandering.
Widescreen mode for the DeepSeek Harness session detail page: a General-settings toggle that lets the chat use the full browser width instead of the fixed 748px content column.
dsh 插件：给文本模型补一双眼睛 + 给多模态模型补一对观察 UI 渲染的眼睛 —— 设置弹窗选视觉模型 + meow_vision 工具 + meow_preview 组件截图工具
手机端 UI 适配：窄屏下把 DSH Web GUI 重构为全屏聊天布局（侧栏/右侧面板改为抽屉，顶部悬浮操作条 + 全屏按钮），默认 ≤860px 生效
One-click desktop launcher for DSH Harness: automatically installs a prebuilt native app (DSH Harness.exe, whale icon) on the desktop when the plugin activates (parameters auto-detected from the environment); the app starts DSH hidden and opens the GUI as a fullscreen Edge app window, and leaves a tray-resident whale with live server status, one-click server restart and optional boot autostart. The sidebar 鲸湾 (Whale Bay) menu (Reload / Restart server / Full exit) renders as a plain sidebar.footer.action neighbor via React portals — no DOM relocation, no hidden native triggers — so it coexists with every other footer-action plugin (including dynamic Cordis plugins) and follows the harness display language; a dsh-dock settings page (tray residency / autostart / extra DSH install paths) lives in the Settings dialog. Candidate discovery is adaptive: it auto-scans every global install the machine exposes — the running Node, roaming npm, WinGet Node versions and nvm-windows version dirs — by probing each unique node's `npm root -g` (v0.6), and lets the user add any custom install path from the settings page (v0.7), so no one ever has to relocate a differently-installed dsh. v0.8: the tray poll cadence adapts to state (1.5s while booting/restarting, 15s steady) and a file watcher delivers the sidebar restart request, Full exit finishes as soon as the host signals disposal instead of waiting out a fixed watchdog, and a restart request is consumed exactly once.
DeepSeek Harness plugin that turns UI screenshots into implementation-grade web specs using OCR, deterministic geometry, scene graphs, assets, and render comparison.
DeepSeek Harness bundle for a Windows desktop UI context picker and MCP server for Codex, DeepSeek Harness, and AI agents
SSiD (思灵) quick screenshot capture: tray/hotkey → fullscreen box-select overlay → cropped image into the current conversation composer
Paper Design MCP bridge as a DeepSeek Harness bundle: paper_* tools, image screenshot support, design skills, OAuth auth, and Cursor-parity design guidance.
Select a range of a DeepSeek Harness conversation and share it as a branded PNG long image, with draggable magnetically snapping range markers.
A terminal dock merged into the DSH web GUI: hideable, split-screen with the chat column, and integrable with terminal-designed TUI plugins (host registerView service + window CustomEvent channel). Windows ConPTY via node-pty, xterm.js in the browser, WebSocket bridge with backpressure.
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.
Vision + Android UI automation plugin for DeepSeek Harness: gpt-5.6-luna screenshot understanding and real-tap mobile app testing via adb