dsh-vision-opencode
Adds a configurable vision model to text-only main models: a vision_read_image tool, a composer-bar vision-model selector, and automatic image-to-text conversion for text-only routes.
给纯文本主模型加可配置识图模型：vision_read_image 工具、输入框识图模型选择器，以及纯文本路由的图片自动转文字。
How to install
dsh plugin add github:poiuyjie/dsh-vision-opencode About
dsh-vision-opencode **中文** ｜ English *点击上方链接切换语言 · Click the links above to switch languages* **DeepSeek 不认图怎么办？OpenCode 这套多模态平替方案了解一下。** DeepSeek Harness（DSH）插件：给纯文本主模型加一个**可配置的识图模型**。主模型（如 DeepSeek）不认识图片？没关系——图片先交给视觉模型（如 MiMo-V2.5）理解成文本，再把结论交给主模型，全程不用换掉你正在用的主力模型。 输入框右侧「识图模型」下拉选择器（自动列出所有供应商中声明了图片输入的模型），样式与交互复刻 DSH 官方模型选择器（含分组标题、尾部对勾、方向键导航、失败重试与 Toast），触发器带「Vision」前缀 运行时 skill vision-image-analysis：按需指导主模型调用 vision_read_image 做 OCR / 图表 / 场景理解；未启用 DSH skill 服务时自动退化为工具与系统提示词 用户在聊天里发图片时**自动转换**：识图进度临时显示在选择器触发器的「Vision」槽位（运行中为官方同款微光动效），结束后只用 DSH 原生 notice 留下一条完成或失败记录 自动识别当前适配器目录中的纯文本主模型并接管图片转换；原生…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 13
- mo. downloads
- –
- Last push
- 2026-09-09
- Created
- 2026-08-13
Links
Basic safety check
- Findings
- curated 收录但无 npm 包/安装命令
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh, dsh-plugin, dsh-plugin-market, dsh-plugins
Related plugins
modlens
liustack/modlens
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
dsh-web
zhu1090093659/dsh-web-ui/tree/main/packages/dsh-tool-describe-image
A `describe_image` vision tool for text-only models: images (local path, URL, attachment) go to a configurable OpenAI-compatible vision endpoint and only the returned text enters the session.
dsh-vision-router
ysr666/dsh-vision-router
Free vision for text-only agents: built-in keyless vision chain plus pixel tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots); paste an image to use it.
dsh-vision-toolkit
Anionex/dsh-vision-toolkit
Vision for text-only models: paste an image and the model switches to a Vision Toolkit variant for image Q&A, multi-image comparison, long-screenshot OCR, screenshot-to-UI reproduction, element grounding, and pixel diff. No API key by default — images are processed by the author-hosted free service, 100 per machine per day; configurable to your own provider.