Vision tools · DeepSeek Harness plugins — Page 44

dsh-FreeBackground — a WebUI background customizer for DeepSeek Harness: upload images or a folder, single-image or slideshow mode (sequential/random), image/UI opacity and blur controls, with light & dark theme support.

前端设计技能：指导 agent 产出高质量 Web 前端设计（视觉层级、配色、排版、组件、响应式、无障碍）。改编自 Anthropic 官方 frontend-design（Apache-2.0）。

Upload an image as a full-window frosted-glass theme for DeepSeek Harness Web. Official Light / Dark / System stay in charge.

DeepSeek Harness 插件：为 DeepSeek 模型桥接 Google Gemini 的多模态识别与图像生成。注册 gemini_vision / gemini_generate_image / gemini_optimize_image 三个工具，设置页填写 API Key 并自动/手动选择视觉与生图模型，生图后自动识别检查并迭代优化。

DeepSeek Harness plugin: multimodal tools (image/audio/video/document understanding, transcription, image generation) via Gemini API or the local Antigravity CLI.

Multi-account Google Gemini / Antigravity / Cloud Code Assist provider for DeepSeek Harness

格式塔视觉感知原则参考

DSH plugin that drives a local macOS camera gesture mouse helper.

Custom background (images / GIFs / animated wallpapers) plugin for the DeepSeek Harness (DSH) Web GUI: an enable switch plus a local asset library, with separate display mechanisms for the official theme and skin-center skins.

Glassmorphism UI theme for DeepSeek Harness web: frosted-glass surfaces, adjustable opacity/blur, custom fonts, background image & live wallpaper with corner/app/fullscreen modes, and soft fade-in animations. DSH Web 毛玻璃 UI 主题：可调透明度/模糊、自定义字体、背景图与动态壁纸（左下角/整界面/全屏三模式）、柔和渐显动画。

GLM 视觉模型插件：注册 glm-vision 供应商路由（glm-4.6v 系列，image+text 输入声明）并提供 glm_vision 工具，让 DeepSeek 等文本主模型直接调用智谱视觉模型看图。

DSH GUI Toolkit — Windows desktop control tools for DeepSeek Harness (DSH): screen layout, annotated screenshot, single-window capture, click / drag / type / key / scroll, semantic verification, window management, and UI Automation structured queries. Twelve model tools, zero external dependencies, no resident service.

Local OCR for DSH: handwritten Chinese and math formulas to Markdown with LaTeX. GPU (DirectML) / CPU / NPU backends, settings-driven, one-click install.

Application-level Vision-Language-Model (VLM) analyzer for DeepSeek Harness: analyze_image tool with primary/backup OpenAI-compatible endpoints, automatic failover, and an auto-saving web settings page.

Tencent Cloud Hunyuan 3D (混元生3D / ai3d OpenAPI) modeling tools for dsh: text-to-3D and image-to-3D job submission with result polling and on-disk asset download.

Custom desktop icon launcher: preset icons plus arbitrary-image to .ico desktop-shortcut creation, with a hosted settings card (host + browser halves)

DSH 插件全家桶聚合包：一条命令从 GitHub 装齐界面增强 / 视觉与工具 / 输入与搜索 / 技能与工作流 / 历史与状态。

图片处理：读取图片尺寸/格式、缩放与格式转换（内置 ffmpeg，跨平台）

Keep native vision, drop historical images before they hit relay providers. Global DeepSeek Harness bundle for every agent.

DSH web plugin: click a pending image in the composer to annotate it (pen, box, arrow, text) and insert the annotated PNG back as a new attachment.

Image auto-describe provider for the DeepSeek Harness apiproxy admission seam: transcribes pasted images through configurable vision routes (Qwen first, GLM fallback) so text-only session models still receive image prompts.

DeepSeek Harness plugin: transcodes images to a base64-safe format (PNG) before a vision request, fixing gateways like LM Studio that reject non-PNG/JPEG/GIF data URLs with "'url' field must be a base64 encoded image."

Give text-only models eyes: an analyze_image tool for DeepSeek Harness, backed by free Chinese vision APIs (GLM-4V-Flash / Qwen-VL) or any OpenAI-compatible vision endpoint. 给纯文本模型装上眼睛。

让文本模型（如 DeepSeek）也能正常粘贴并发送图片：图片落盘到工作区，消息转成仅模型可见的路径文本，由模型调用视觉/MCP 工具识别。