60 results
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.
DeepSeek Harness plugin: auto-detect secrets pasted into the composer, store them in the official credentials seam, and send [secret:REF] placeholders to the model instead of the value.
Paste images / drag any files into the DSH composer -> saved into the session workspace as files -> images read by modlens, other files read by the model by path
Plug-and-play image reading for text-only models: admits pasted images, projects them as [Image #N], and provides a first-class read_image tool backed by a configurable vision model, with a 读图 settings page (设置 → 读图) bridged over typert Remote.
快捷附图：在会话输入框粘贴图片时自动保存到用户配置的目录，发送消息时把图片的绝对路径一并发送给 AI 识别。设置栏中显示为「快捷附图」。
VS Code-style file explorer plugin for the dsh web GUI: file panel beside the sidebar, multi-tab preview panel (code/images/PDF/audio/video/Markdown/CSV/Excel/Word/zip), filename search, sort/filter, copy-paste, swap sides and drag-resize — zero patches to official packages
DeepSeek Harness plugin: drag-and-drop file attachment — paste or drop any local file into the DSH web composer as an @workspace path reference
dsh auto-vision bridge: paste an image into a text-only model's composer and a configured multimodal model transcribes it as text, automatically.
Vision recognition plugin for DeepSeek Harness: paste images into the composer, recognize them via GLM-4V on the host side, and inject the result into the conversation.
自包含的 DeepSeek Harness 视觉系统：DeepSeek 视觉孪生路由（原生图片体验 + 视觉桥请求层拦截）+ 本地 Ollama Agentic Vision 工具（describe/OCR/结构化扫描/区域查询/元素定位/双图对比/剪贴板）+ 粘贴分流（paste-to-path）。安装即用，无本机路径依赖。
dsDragPasteALL：拖拽任意文件/粘贴非图片文件到 DSH 页面（纯图片粘贴交给原生成图），后台静默反查文件真实路径（Spotlight + find + 首尾哈希校验），命中后把路径直接插入对话框输入框；找不到原文件则内容缓存兜底。
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.
Custom background wallpaper for the DeepSeek Harness chat page — upload a local image or paste an image link, tune opacity and brightness, scope to chat area or full screen.
Plugin manager for the dsh web GUI: paste a GitHub repo URL, verify it is a dsh plugin (bundle or pure cordis), install, then update / remove / enable / disable across the full lifecycle, with background auto-update.
DeepSeek Harness (DSH) host-layer plugin: detects and auto-fixes Windows PowerShell command syntax mistakes — bash constructs, broken line continuations, pasted prompts, CRLF pollution — then executes the corrected command through the host shell seam under the normal sandbox/approval policy
File upload for the DeepSeek Harness web GUI: a folder button, paste, and drag-drop that save files into the session workspace uploads/, with file cards shown alongside image thumbnails.
deepseek-eyes：给 DeepSeek Harness 的纯文本模型装上眼睛——把粘贴的图片经 OpenAI 兼容视觉 API 转成文字描述，再转发给真实的文本模型（Give text-only models in DeepSeek Harness eyes: describe pasted images via an OpenAI-compatible vision API, then forward to the real provider. Not affiliated with DeepSeek.）
On-demand vision for text-only LLMs in DeepSeek Harness: keep pasted/attached images in the backend, and let the model call view_image to look at them anytime via any OpenAI-compatible vision endpoint — as if the model were natively multimodal.
DSH 文件适配插件（面向 DSH 0.1.2-alpha.1）：拖拽/粘贴/点击上传/@ 文件引用加入本地文件，以「文件：<路径>」引用随消息发出；主模型按需调用 dshomnifile 工具解析——文本多编码直读、Office/PDF 等文档由 anydoc 提文本、文档内嵌图片与扫描/图文混排 PDF 页由进程内 pdfjs 渲染 + 配置的多模态模型识别（内容哈希缓存 + 并发排队）并组装为 Markdown；聊天内文件卡片可预览图片、本地打开源文件；识别走官方 ctx.llm.stream。
Image routing for text-only models in DeepSeek Harness: a global analyze_image tool (Kimi vision) plus automatic rewriting of pasted images into attachment references when the active model cannot see images.
Drag and paste files into dsh Web with a unified preview rail; keep native images and add file references only when sending.
Paste or drag non-image files (xlsx/csv/pdf/zip…) into the DSH web composer: files are saved into the session workspace and an @reference is inserted.
DeepSeek Harness plugin: upload/paste images in the chat box; on send, transcribe via a vision model (dashscope) or offline Windows OCR, then inject the description into the message as 【解析了提供图片，图片内容是<描述>】.
Screenshot/file paste board for the DSH Web GUI: paste, drag or browse to save files locally, show reference chips above the input, and auto-append file paths when sending a message.