視覚ツール · DeepSeek Harness プラグイン — 15 ページ

One-key ClawRouters plugin for DeepSeek Harness: chat, image generation, video generation, and web search

Local image/voice/music/SFX for DeepSeek Harness with asset consistency: the same character stays the same character across calls, degenerate output is refused, and the GPU is untouched when idle.

Upload an image as the whole-page background wallpaper with adjustable dark scrim / opacity / blur, surfaced in Settings → 自定义主题

Adds opt-in image-capable sibling routes for text-only providers and converts image blocks into budgeted, locally enhanced OCR, color-statistics, pixel-scan, and metadata evidence before delegation.

Qwen-MM-Plugins integration for DeepSeek Harness: 12 MCP multimodal tools (vision/OCR/grounding/ASR/AV), a Web settings page (paste a Qwen API Key), bundled skills, and a one-command installer.

DSV4.1 reasoning support with multimodal context and bounded artifact acceptance for DeepSeek Harness

Show a local image or short video inside the current DeepSeek Harness conversation card, with click-to-preview. Does not replace the official read_image model path.

Plug-in vision for text-only DeepSeek Harness (dsh) models: a `vision` tool with built-in cheap/free VLM presets, multi-image batch analysis, paste-to-hint image admission, and a web settings page with hot-reload.

DeepSeek Harness appearance skin: minimal themes, accent/text/region colors, image skins, named presets, durable settings.

DSH 视觉原语工具：参考 DeepSeek《Thinking with Visual Primitives》论文，将图片路由到外部视觉模型并返回带视觉基元的文本分析。纯文本循环，对话模型无需原生视觉能力即可'看见'图片。

Browser automation toolset for dsh-tui agents: Playwright-driven browser control with deepseek-v4-flash-vision-exp visual understanding.

Turbo Vision-style character-cell windowing IDE for DeepSeek Harness agents

Customize the DeepSeek Harness web UI: theme color presets (main area, sidebar, message bubble, send button), full-page background image, plus a VSCode-style card layout engine (dock/undock/float panels, tabbed dock host) — all reversible and light/dark aware.

Full DIY theme plugin for DSH web: color palettes, fonts, background image (upload/URL), glassmorphism, border radius.

DeepSeek Harness plugin that turns UI screenshots into implementation-grade web specs using OCR, deterministic geometry, scene graphs, assets, and render comparison.

Veneer — a lightweight visual editor to reskin the DeepSeek Harness UI: edit background surfaces (color/image/opacity) with theme persistence.

给纯文本模型自动配眼:图片消息自动桥接为视觉模型描述(日志内表面替换,不改写人类转录),附带 see_image 追问工具

DeepSeek Harness (DSH) 识图插件：为不具备原生识图能力的模型提供识图能力，把图片发送给阿里云百炼/智谱多模态模型并返回文字描述。由 claude-vision-skill 移植适配。

Local vision bridge for DeepSeek Harness

DeepSeek Harness 全能插件：识别/生图/改图一体化，无需切换模型——用常规 DeepSeek 模型即可自动调用视觉与生图模型（gemini_vision / gemini_generate_image / gemini_optimize_image）。多后端：Gemini 原生 + 任意 OpenAI 兼容服务（GPT-4o、Qwen-VL、GLM-4V、Moonshot、gpt-image、DALL-E、Flux、Stable Diffusion、OpenRouter、硅基流动、各类中转等），生成后自动视觉自检反馈，优于 modlens。

A DeepSeek Harness Mix plugin for vision routing plus GPT Image generation and editing.

DeepSeek Harness adapter that exposes each configured vision model as a selectable DeepSeek composite in Web UI.

DSH standard vision skill: Qwen dynamic-resolution preprocessing + OpenAI-compatible VLM chain with failover/429 backoff, structured evidence mode, local tesseract-first long-screenshot OCR, paste-to-path (no framework patch). 8 tools + runtime skill.

vqa_ask 双模型视觉问答插件:主模型调用 vqa_ask 向视觉模型提问(图片字节真实送达),UI 实时展示"主模型提问 → 视觉模型回答"的 QA 过程,并在主设置提供多模态视觉模型选择页。