vision 117

dsh bundle: subagent_vision — delegate image reading to a vision-capable model from a text-only session, plus paste-to-path so pasted images reach the subagent as file paths.

Conversation-style UI/UX visual review plugin for DeepSeek Harness: vision_review / vision_ask tools backed by any OpenAI-compatible multimodal endpoint (e.g. agnes-2.5-flash)

Native-vision capability manager and lightweight external vision fallback for DeepSeek Harness

Lets a text-only agent call a multimodal model mid-task: the vision tool sends one image file to Qwen, Kimi, OpenAI, Claude, Gemini or a self-hosted endpoint and returns structured evidence — summary, verbatim OCR, layout, entities, and what the model cou

Describe images with a vision model so text-only DeepSeek Harness models can read them

mm-vision (通感编码器) for DeepSeek Harness — give any text-only LLM the ability to see images via structured spatial text encoding. Registers the mm_vision tool.

DSH 插件：通过硅基流动(SiliconFlow)视觉大模型识别/分析图片，支持本地文件路径、http(s) 图片 URL 与 base64 data URL。含持久化的粘贴识别面板（web 端）。

Model-facing accurate_vision tool: precise image spatial reasoning via a vision model

DSH plugin: image understanding via any OpenAI-compatible vision API, paste-to-describe, and an animated whale-buddy desktop pet with status bubbles and a floating settings panel

DeepSeek Harness plugin: content-aware PDF reading tools for vision models (pdf_scan / pdf_read_page / pdf_render_region). Detects figures (vector+raster), tables, math, and two-column layout per page, then routes text pages to extraction and figure/table

DSH 图片插件，三个工具覆盖三种场景：ask_user_choice 图片/图文混合选择卡（Web GUI 渲染，可放大查看）+ show_images 回复内嵌图片（图文混排）+ save_received_images 盲模型收图存为工作区文件；聊天栏所有图片点击放大，支持滚轮缩放与拖拽平移。来源支持本地路径 / http(s) URL / base64 data URI。零 token 本地渲染，纯插件实现不改核心包。

Paste images into DeepSeek Harness chat and have them read by a free local backend (macOS Vision / Tesseract) before the text-only DeepSeek model answers

给 DeepSeek Harness 加视觉能力：可视化设置页选视觉厂商（Kimi/智谱/通义/OpenAI/Gemini/豆包/MiniMax/阶跃星辰）并粘贴 API Key，聊天里即可拖图识别——图片走视觉模型、文字走 DeepSeek 推理。

Transparent image guard + vision analysis for DeepSeek Harness: text-only models read pasted images without the 400 session deadlock.

Persistent vision plugin for DeepSeek Harness: registers the vision_analyze tool backed by a configurable multimodal model, with a settings-page UI. Zero dependencies.

模型能力管理：为手写模型设置图片输入与推理强度、线路系统消息角色（可一键继承目录模型的精确能力）。

Local-first vision for text-only DeepSeek Harness agents: route image understanding to a local OpenAI-compatible vision model, return structured JSON evidence (summary / OCR / layout / semantics / visual / uncertainty).