Vision tools · DeepSeek Harness plugins — Page 49

Declare whether a configured third-party model accepts image (multimodal) input; writes the modality into the owning provider settings and verifies it through runtime model resolution.

DeepSeek Harness plugin to audit and correct llm-pi-ai model capability declarations by probing the endpoint itself: out-of-range maxTokens, contextWindow, reasoning levels, image input.

One-click model + reasoning-effort switcher for the DeepSeek Harness web UI: capsule trigger, gradient effort slider with detents, and a premium MAX-state visual.

DSH 增强功能静态化常驻包(bundle):余额查询/会话完成提醒/费用预估与账单明细/粘贴图片转文字/Token 用量统计;仓库另含模型选择器增强等插件源码与一键安装脚本

DSH WebUI 动态背景 —— 可插拔效果内核。效果以 mod 形式放在 mods/ 目录，放进文件夹即多一个效果。内核为原创代码，不含任何第三方着色器

WeChat Mini Program automated testing for DeepSeek Harness (dsh): selector-addressed actions, build-freshness gates, geometry-first assertions that work on text-only models, and screenshot image blocks on vision routes. 微信小程序自动化测试 dsh 插件：选择器寻址、构建新鲜度门、面向纯文本模型的几何断言、视觉路由附真实截图。

DeepSeek Harness plugin: MxPage ecommerce product-image pipeline (analyze → plan → VPA → generate).

Native zoom, pan, download, gallery, and region-note image viewer for DeepSeek Harness

网易云音乐工具插件：单曲（含无损/母带）、MV、.lrc 歌词、云村笔记、歌手/用户主页、图片原图、评论/排行榜/听歌排行，歌单查看与增删改，以及 AI 私人电台策展。

Call a new-api (one-api style) relay station's video model from the conversation, with @-referenced image/video materials plus a prompt.

DSH sidebar notebook: notes with a title and a body, image attachments (videos rejected), one-shot clipboard copy of the body, and a single reusable editor dialog.

DSH plugin: local OCR and image understanding for dsh — inlined image previews in Web tool cards

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

DeepSeek Harness OCR 插件 — 基于小笨羊 OCR 能力，为纯文本模型提供图片文字识别功能

Self-healing guard for the dsh-ocr plugin: re-applies the llm-deepseek adapter OCR seam (and the dsh-ocr package fixes) after a dsh upgrade wipes them. Fail-safe: verifies the patched module loads, rolls back on failure.

Direct DeepSeek official vision API bridge for DeepSeek Harness: registers the deepseek-v4-flash-vision-exp multimodal model as a provider route, with base64 inline and Files API image input. 直连 DeepSeek 官方视觉 API 的 DSH 插件

DSH vision bridge (DSH >= 0.1.2-rc.1): when the selected chat model is text-only, attached images are described by a local Ollama VL model (qwen3-vl:8b) with keep_alive VRAM cooling. Install-time patch of dsh-api-session-controller prompt admission + runtime status companion.

A canvas the agent draws on and then sees: eyes_render draws text/shapes in the Web GUI, stores the PNG locally, and hands the result back to the model. Windows-only: eyes_ocr reads text via the built-in Windows OCR engine (Windows.Media.Ocr).

DSH 文件适配插件：拖拽/粘贴/点击多选本地文件，anydoc 解析文档，多模态模型识别图片，整合给主模型。

给 dsh 加上 QQ 通道（OneBot 11，兼容 NapCat / Lagrange / LLOneBot）：反向/正向 WS 自动重连、私聊群聊、图片/语音转写、t2i 文字图卡片、合并转发、设置页可视化配置。QQ channel for dsh via OneBot 11.

Alibaba Open Code Review for DeepSeek Harness: native tools and a host-model review skill, with optional OCR-managed reviews.

Community OpenAI account model connector for DeepSeek Harness

OpenRouter provider for DeepSeek Harness with a live model list: every model OpenRouter serves, with context sizes, reasoning levels and image input read from its /models endpoint.

DSH DeepSeek Harness plugin: Add PaddleOCR vision to text-only LLMs (Agnes, DeepSeek, MiniMax, GLM, Mimo). 为DeepSeek Harness纯文本模型添加图片识别能力，支持本地截图和文档OCR。