視覚ツール · DeepSeek Harness プラグイン — 52 ページ

DSH (DeepSeek Harness) file preview & editor plugin: in-browser workspace file tree, multi-format preview (code, Markdown, PDF, image, Office, HTML), online editing with version history, and full-text search. DSH 文件预览插件：工作区文件树浏览、代码/Markdown/PDF/图片/Office/HTML 多格式预览、在线编辑与版本回滚、全文搜索。

dsh-std Community v0.15 ModelProvider: DeepSeek-V4-Pro with images captioned by V4-Flash-Vision-Exp. Requires @dsh-std/adapter-dsh on DeepSeek Harness.

DeepSeek Harness bundle that refuses shell commands terminating browsers or the harness's own host processes by image name.

Custom provider enhancements: User-Agent override, reasoning-level switching, and image-input declaration for DSH custom providers · 自定义供应方增强：UA 覆盖、推理等级切换、图片输入声明。

Bridge QQ official bot private messages (C2C) into DeepSeek Harness (dsh) agent sessions

QQ Bot (QQ 机器人) IM channel plugin for DeepSeek Harness (dsh): per-peer sessions, group @-mention and keyword triggers, image generation, TTS voice replies, attachment handling.

快捷附图：在会话输入框粘贴图片时自动保存到用户配置的目录，发送消息时把图片的绝对路径一并发送给 AI 识别。设置栏中显示为「快捷附图」。

QuickLook-style space-key large preview for dsh-better-sidebar: press Space on the active file tab for an instant image / PDF / text overlay

Two-tier image reading for text-only models in DeepSeek Harness: fast local OCR (RapidOCR, offline) first, then a vision model via modlens as fallback.

DSH bundle: Qwen multimodal bridge — vision (qwen3-vl), speech-to-text (qwen3-asr), text-to-image (qwen-image), via the deepseek-vision skill scripts

Dual-face (host + client) plugin: independent text-output ocr_image tool + tools/execute read_image capability dispatch + ctx.ocr seam, with a conversation.input.left OCR toggle button.

Render read_image tool results as real images in the DeepSeek Harness Web GUI: a dedicated Read image row (thumbnail with original-image lightbox, retry) plus the metadata envelope, and the same image in the Inspect details panel. 在 Web GUI 对话流中把 read_image 的结果渲染成真实图片。

DSH profile plugin: converts read_image PNG/WebP attachments to opaque sRGB JPEG so LM Studio's openai-completions endpoint accepts them

Show local images read by read_image directly in the dsh Web chat

自适应 GitHub README 撰写与截图技能：先判断目标类型与截图能力——网页走浏览器截图，无网页的控制台/CLI 走终端截屏（System.Drawing / HTML 终端 / termtosvg），能则询问并自动配图，不能则走无截图模板；强调相对路径防丢图。作为 dsh 插件（cordis skill-bundle）安装后由 host 层注册 readme-writer 技能；SKILL.md 含可被插件商城索引的 name/description/keywords/category/version 元数据。

Media library for the dsh web UI: browse configured local directories as a grid or list, open images in a viewer, and play videos in a page-fullscreen player with swipe navigation, order/random playlists and a fullscreen wallpaper.

Reference-image search for the dsh web GUI: a model-facing search_refs tool that queries Openverse (keyless) or Pexels / Pixabay / Unsplash (API key) for design reference images, plus a browser half that renders the results inside the conversation as a Pinterest-style masonry wall with a lightbox, a dsh-better-sidebar board tab, one-click batch/page controls, proxied downloads, and save-to-Eagle.

DSH rembg plugin with CPU/GPU background removal and setup UI

🚀 告别繁琐 SCP！远程部署 DSH 一键下载修改后的文件与图片预览交付插件

DeepSeek Harness (DSH) 模型输出复读熔断插件：P1 通用流式复读检测（llm/stream 在线拦截）+ P2 Vision Toolkit 局部防护，命中即终止流并给出结构化错误码，避免模型复读浪费大量 output tokens。

DSH plugin: free web search tool (DuckDuckGo via ddg-kit with system-proxy support, automatic Bing fallback) + front-loaded vision understanding tool (vision_understand: understand images via a vision-capable model configured in DSH when the conversation model has no image modality) + composer-level paste-image capture (paste/drop an image in the conversation input → vision model → description, bypassing DSH's image gate on text-only models) + drag-drop file stash (drag/paste/attach any file → stashed into the session workspace, attachment list injected into the next message via agent/pre-step — clean draft, replayable history) + token auth toggle (turn off token verification for open access, effective immediately, persists across restarts). Official bundle plugin, install: dsh plugin --profile web add github:LingyeSoul/dsh-rider#main。DSH 插件：免费网络搜索（duckduckgo_search 工具）+ 前置视觉理解（vision_understand 工具）+ 对话输入框粘贴图片捕获（粘贴/拖入图片 → 视觉模型 → 描述，绕开 DSH 对纯文本模型的图片拦截）+ 对话窗口拖拽上传任意文件（拖入/粘贴/选择文件 → 暂存进会话工作区，附件清单随下一条消息经 agent/pre-step 自动注入，草稿零污染）+ Token 验证开关（关闭后免 token 直接访问，立即生效，重启保持）。

DSH web plugin: a right-docked file panel that browses any directory and views/edits text, images (brightness/contrast/grayscale/rotate/flip), and Office/PDF documents (.docx/.xlsx/.pptx/.pdf via local Python).

Role-play conversations, character-card authoring and image generation for DeepSeek Harness

ROS2 debugging tools and diagnostics skills for DeepSeek Harness (DSH): package/workspace/dependency inspection, node/topic/service/action/param/interface enumeration, topic sampling, TF tree, graph topology JSON, approval-gated builds and message scaffolding, GUI screenshot and multimodal vision observation.