aura-vision
Free vision OCR with adaptive tile recognition for long documents and Markdown/Word/PNG/Excel export.
免费视觉识别插件：自适应切块长文档识别，支持历史收藏与 MD/Word/长图/Excel 导出。
How to install
dsh plugin add aura-vision About
Aura Vision 免费视觉识别插件（DeepSeek Harness web profile 永久插件）。设计语言 **Aura**：宛如天生如此，若有似无。 特性 **免费通道**：智谱 GLM-4V-Flash（免费档）优先；可配置任意 OpenAI 兼容多模态接口；Pollinations 匿名兜底。 **长文档识别**：glm-4v-flash 输出上限 1024 tokens（API 实测）——自适应网格切块（单块目标 1100px、最多 3×3、8% 重叠）逐块识别突破上限，切块模式用逐块纯转录提示词；>3200px 温和归一。 **Aura UI**：高透毛玻璃（55% 底色 + 28px 模糊）、三星机身式小圆角、深浅主题自洽配对（CSS light-dark()）、若有似无的次级信息。 **历史**：缩略图 + 原图分离存储；收藏、过滤、删除、清空；导出 Markdown（原图内嵌 base64，单文件自包含）。 **结果导出**：MD / Word(.doc) / PNG 长图 / Excel（含表格时）；双击预览图与详情图全屏放大；清除图片联动清除结果。 安装（永久） sh 推荐：npm 安装 dsh plugin --profile web add aura-vision 或从 GitHub： dsh plugin --profile web add …
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 1
- mo. downloads
- –
- Last push
- 2026-08-20
- Created
- 2026-08-19
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh-plugin, glm-4v, ocr, vision
Related plugins
modlens
liustack/modlens
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
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.
dsh-deepseek-vision
siegfly/dsh-deepseek-vision
A vision-language gateway provider route: pasted images are described by a configurable VL model (Qwen-VL by default) before the DeepSeek wire.