시각 도구 · DeepSeek Harness 플러그인 — 42 페이지

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

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).

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 创意指导 Agent 插件：6 维评估 + 自适应追问生成通用美学 brief（JSON，平台无关），AIGC 痕迹去除润色，框架逻辑抽成可测试工具（cd_assess / cd_detect_conflict / cd_polish_type / cd_save_brief）

DSH × Creative Studio — a portable AI image-retouching and video-editing skill pack for DeepSeek Harness: background removal with quality tiers, inpainting retouch, text/logo compositing, batch platform exports, footage review, storyboards and timeline editing. · 美工 + 视频剪辑技能包：抠图/改图/文字/logo/合成/批量规格化 + 素材通览/脚本/剪辑/字幕/配乐/封面。

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.

Replace the DSH top-left BrandWordmark/FishLogo with your own image or text — an inline ⚙ handle inside the wordmark, no floating widget

dsh web Settings section that edits per-model image-input support and reasoning-effort mapping, plus dropdown-driven subagent delegation-model config (static agentOptions + per-delegation allow-list).

DeepSeek Harness Web GUI 个性化：自定义侧边栏 Logo、空对话欢迎区（Logo + 文案）与对话背景图，修改实时生效、无需重启

Local office document parsing (PDF/DOCX/XLSX/PPTX) with citation refs, plus optional VLM OCR, for the DeepSeek Harness (developer preview)

Auto-switch DeepSeek models for DeepSeek Harness: routes every model request to the vision model when images are present, to the pro model for complex tasks, and to the fast model otherwise; includes a switch_model tool for manual per-session overrides.

DSH plugin (DeepSeek vision bridge): give text-only agents vision through chat.deepseek.com's unofficial vision mode — WeChat-QR/password login in Settings, a deepseek_vision model tool, and automatic in-chat image transcription.

设计风格参考：主流视觉风格（极简、拟物、玻璃态等）特征与选用建议（Design style reference for DeepSeek Harness）

DSH host plugin: OS-level desktop control tools (screen OCR with coordinates, mouse click/drag, Unicode typing, key combos) so the model can operate the whole browser including the address bar, and any other window.

macOS desktop control for DeepSeek Harness: agent operates non-browser apps (CapCut/PS/WPS/native clients) like a human — see screen, move mouse, type text. 14 tools + vision closed-loop (see/locate/click/verify).

官方 dsh 桌面壳。自带视图给 DeepSeek 带上眼睛；内置锚定模式相对 Standard 约 +8%（Project2 Ability 91→98/99）。

DeepSeek Harness 桌面版（Windows）· Electron desktop shell for DeepSeek Harness — 双击即用 / double-click to run · 支持选区截图提问、系统托盘、会话管理 / region screenshot, system tray, session manager

DMXAPI-DSH配置工具：在 DeepSeek Harness Web 新增服务商，并直接编辑已有服务商的模型。

Local PDF, Office, image, and OCR document intelligence for DeepSeek Harness.

DeepSeek Harness plugin: Docker container, image, log, and Compose tools for the agent, over the local Docker CLI

Project PDF library with background indexing and cited page highlights for DeepSeek Harness Web

Drag non-image files/folders into DeepSeek Harness web and insert their real absolute paths into the composer — no more "only images allowed". 拖拽工程文件/文件夹进 DSH，自动定位真实绝对路径插入输入框。

Extend the DeepSeek Harness (DSH) web chat drag-and-drop to accept every file type: dropping any file in the composer saves it into the current session's workspace so the agent can read and use it.

拖入本地文件时，自动把文件在电脑上的绝对路径填入输入框并复制到剪贴板 ｜ Fill a dropped local file's absolute path into the DSH composer