Vision tools · DeepSeek Harness plugins — Page 11

Drag-and-drop file and folder attachments for the DeepSeek Harness Web UI

Clean desktop shell for DeepSeek Harness (DSH) as a DSH plugin — wraps your web profile in a native window, tray-managed backend, offline auto-reconnect, zero visual changes. DSH 插件形态的纯净桌面壳：复用现有 web profile，托盘管理后端，零视觉改造。

Customizable brand area for the DeepSeek Harness web GUI: replace the whale logo and the DeepSeek wordmark with local images, and edit the HARNESS badge text — double-click to change, right-click to reset. 自定义 DSH 左上角品牌区：logo 与 DeepSeek 文字可换本地图片，HARNESS 徽章文字可双击编辑。

AI-assisted Minecraft for DeepSeek Harness: drive an autonomous Mineflayer agent (survival, mining, chat, live map) with vision/control tools, plus a built-in launcher (Microsoft sign-in, version download, game launch).

Free image reading (vision) and image generation skills for DeepSeek Harness — Zhipu GLM-4V-Flash (Google Gemini optional) for reading, SiliconFlow Kolors for generation.

Medical imaging specialist tools for deepseek-harness: MAIRA-2 and MedGemma report generation for chest X-ray/CT/MRI/retinal (fundus) images, TotalSegmentator organ segmentation, BiomedCLIP zero-shot ultrasound classification, and BiomedParse text-prompted segmentation across X-ray, CT, MRI, ultrasound, and retinal (fundus) images.

Xiaomi MiMo search + multimodal tools for DSH agents (mimo_search/vision/audio/video/asr/tts) plus the merged voice UI: 🔊 read-aloud, in-conversation speech strips, and a configurable voice map (docs/adr/0001).

Local OCR fallback for DeepSeek Harness (Web): when the routed model cannot accept image input, a pasted image is saved locally and its text read by PP-OCRv5 + ONNX Runtime — fully offline, no vision model required. / DeepSeek Harness 本地 OCR 兜底插件（Web）：当接入的模型不支持图片输入时，自动把图片存到本地并用 PP-OCRv5 读出文字，完全离线。

Markdown campus resume workbench: the agent edits resumes and CSS templates under jobhunt/, and the user previews and exports from the sidebar.

DSH 桌宠 bundle 插件（v2.3）。住在 Web 界面右下角的精灵宠物 + 一个独立于浏览器的原生桌面窗口：随 Agent 工作状态切换动作（对话中专注工作、回合思考、等待审批、出错摔倒），对话完成时跳跃庆祝并全机播放提示音。原生桌面窗口最小化/关闭网页也照样存在，悬停显示当前工作区与对话名，双击用 Edge 打开 Harness 网页。内置多宠物注册表（跨重启保留）、每宠物独立音频、素材 HTTP 路由与注册即校验（严格防图片错位）。配套技能 dsh-pet-forge 可一句话生成并导入新宠物（生图模型 + Blender 3D 建模 + 动作/音效设计），也能直接导入视频（抽帧 + 抠绿幕 + 按动作切分）。v2.1 新增宠物社区：按 DPSL-1.0 开放共享协议发布到 GitHub 的桌宠会被自动收录，可在插件内置画廊里浏览、下载、一键安装。v2.3 适配 DSH 0.2 桌面版：修好桌面端庆祝音效（0.2 起默认组合不再提供 ctx.shell，改为 shell 不可用时直接 spawn powershell），并新增单击挑衅动作（pet.json 的 interactions.clickAnim，播完自动回落）。

DSH Tavern 剧情配图插件，支持 ComfyUI、NovelAI、OpenAI Images、Gemini 等渠道。

Lightweight vision assist preset and CLI for DeepSeek Harness.

Wallpaper Engine integration for the dsh web GUI: scan the local Wallpaper Engine library (Steam workshop 431960 + local projects), use its wallpapers as the page background (image / video / web / scene-preview), and control them from a right-side '壁纸设计' panel (opacity, scope, fill, blur, vignette, fps, parallax, carousel, theme linkage). Agent tools wallpaper_scan / wallpaper_list / wallpaper_set / wallpaper_config.

Turn your Wallpaper Engine collection into DSH chat backgrounds. Auto-discovers Steam Workshop and local projects, plays video/image/scene wallpapers live, with glassmorphism UI customisation and carousel auto-rotation.

Renders images, video, audio, PDF and Office documents inline in the dsh web UI through a display_file tool, streaming bytes over a signed HTTP route with range support.

Model-facing describe_image tool: asks a vision-language model at an OpenAI-compatible endpoint to describe an image file or URL

dsh-conv-export（对话导出）— export the current DeepSeek Harness conversation as Markdown, single-file HTML, PDF, or a long PNG image, or copy it as Markdown; batch-export historical sessions as a Markdown ZIP.

A DeepSeek Harness research workspace with archived PDF reading, optional OCR, human review, cited reports, and recoverable research tasks.

Image generation for DeepSeek Harness: a generate_image tool with pluggable providers — the FAL queue, any OpenAI-compatible images API, or a ChatGPT/Grok subscription with no API key at all. The picture is shown inline in the conversation; the model receives either a link (works with any chat model) or the image itself (needs dsh-vision-bridge or a vision-capable model).

DSH Web 皮肤插件：多套品牌皮肤侧栏热切换。

DSH Web plugin that lets text-only models see images: paste images in the chat and send without file paths; the model discovers its own vision tools. Multimodal models pass through natively.

Movable, resizable Chromium desktop embedded in the DeepSeek Harness Web UI through noVNC.

DeepSeek Prism for DeepSeek Harness (DSH): prism_see tool + text-only/image model admission (native pass-through or VEP/2 conversion) + runtime skill + settings card (provider incl. deepseek-vision, deployMode, visionModelHandling). Unified bundle — optional harness patch for original display + fold/progress UI.

Always-on-top DeepSeek Harness desk pet with a native menu and DIY skins. Shows what your agent is doing: working, waiting, finished, failed. Five skins, or make your own from one image. No dependencies.