Vision tools · DeepSeek Harness plugins — Page 10

Moren 无限制梁神模式：把 dsh-purge 的无限制模式/无限制子任务与梁神模式合成一个原生预设，带左侧栏入口、图片与视频识别、本地生图、编码器、Token 守卫和可自检修复的回退。

Display images read by the read_image tool inside the DeepSeek Harness Web GUI: a DSH 0.2-compatible dedicated Read image row with a default-expanded message-style image card (PS-style transparency checkerboard) and an in-page full-resolution lightbox. 在 Web GUI 对话流中展示 read_image 读到的图片：兼容 DSH 0.2 的默认展开图片卡（透明棋盘格）+ 页面内全精度放大层。

网页抓取与分析插件：抓取网页正文并给出统计与关键词分析，支持内容图片下载与识图分析器扩展接口。

DeepSeek Harness 外置视觉模型插件：inspect_image 把本地图片或 http(s) 图片 URL 发给任意 OpenAI 兼容端点，视觉模型看图的文字回答直接带回对话；附 Web UI 设置栏。

DeepSeek Harness plugin: configurable vision model with vision_read_image tool, composer-bar vision-model selector, and automatic image-to-text conversion for text-only main models.

Eyes for text-only DeepSeek on DeepSeek Harness: image understanding via Doubao Web by default (zero cost, no API key), Antigravity IDE quota (flash/pro), Gemini API, or Cockpit proxy — model-invokable vision tool, wrapper adapters, evidence memory, and a bilingual client panel.

一个插件接入全部渠道模型：WorkBuddy（含企业版，多账号注册表）/ Trae / Qoder CN。支持渠道添加、改名、删除、凭据管理、统一状态面板与一键签到。

deepseek-aix right-panel system: a faithful port of the dsh-web-ui aionui-panel — the Explorer file tree with filename search, the multi-tab Preview (markdown/html/code/diff/csv/pdf/office/image/text/url), and the drag-a-file-into-the-composer dock, driven by a workspace-gated host filesystem service over /aionui-panel routes

Pure React attachment atoms for the dsh web UI: draft-image rail, message image gallery, and original-image lightbox (zero cordis)

Pure React attachment atoms for the dsh web UI: draft-image rail, message image gallery, and original-image lightbox (zero cordis)

Pure React attachment atoms for the dsh web UI: draft-image rail, message image gallery, and original-image lightbox (zero cordis)

DeepSeek Harness (DSH) vision bridge: let text-only models delegate image understanding to multimodal vision models via vision_describe tool, with persistent settings UI and primary/fallback auto-retry. 视觉桥：让纯文本模型通过 vision_describe 工具把图片理解委托给多模态视觉模型，带持久化设置界面与主备自动重试。

DeepSeek Harness (DSH) vision bridge: let text-only models delegate image understanding to multimodal vision models via the vision_describe tool, with a lean settings page and primary/fallback auto-retry. 视觉桥：让纯文本模型通过 vision_describe 把图片交给多模态视觉模型，带精简设置页与主备自动重试。

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

Two team Agent presets for the DSH desktop profile: 建模编程 Agent (model-code) and 论文 Agent (paper), with their bundled skills.

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.

Browser automation for DeepSeek Harness: Playwright-powered open/click/type/screenshot/eval tools so the agent can drive real web pages.

Omni-modal workstation plugin for DeepSeek Harness: analyze_image + 6 vision toolkit tools (zoom/sample_colors/image_diff/ocr/detect/show), generate_image, and generate_video (t2v/i2v async-task). Backed by an ordered multi-card list (OpenAI/Anthropic/Gemini protocols; Custom/Ollama plus 28 built-in fixed providers), Tabbed settings (VLM / Image Gen / Video / Mirror / Fallback), single-request failover, mirror model routing, JPEG->PNG re-encode fallback, and an auto-saving web settings page.

[DISCONTINUED 2026-09-11: DSH >=0.1.5 ships a built-in workspace file tree + document preview (dsh-client-ui-sidebar-files / sidebar-documentpreview); please uninstall and use the built-in features] VS Code-style workspace file explorer for the DeepSeek Harness web GUI — editable preview with syntax highlighting, tabs and line numbers, full right-click file operations (create / rename / delete / copy / cut / paste / reveal in file manager), inline image & markdown preview, @-mention files into chat, drag files into the composer, per-workspace state with undo. Turns the dsh web page into a lightweight code editor.

便签静态插件:对话框右侧空白处的便签面板,可打字记录、勾选完成(删除线),localStorage 持久化,随 web 包常驻。

股市行情皮肤+功能插件：从 Trading Terminal 皮肤演进而来的自选跑马灯、实时行情、自选列表与持仓盈亏面板，点击跑马灯弹出个股日K图（蜡烛/MA/成交量）（A股/港股/美股/指数/加密/外汇，腾讯+Binance+Yahoo+Frankfurter 聚合）

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

Drop or paste images, PDFs, office docs, zips, videos and audio into the DSH composer as workspace file paths instead of model attachments — lets a text-only model agent read them via vision tools.

dsh plugin: recognize attached images with the built-in Windows OCR engine (Windows.Media.Ocr) and send only the recognized text to the model. Text models never receive image bytes; vision passthrough is opt-in.