modlens

liustack/modlens

The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding agent. Paste an image, get structured JSON evidence (OCR, layout, semantics). | 全网最强 DeepSeek Harness 外挂视觉插件,为 DeepSeek、GLM 等纯文本模型外挂视觉能力,粘贴图片即得结构化 JSON 证据(OCR、版面、语义)。

Install

GitHub
liustack/modlens
Website
liustack.dev
Stars
★ 4.5k
License
MIT
Language
TypeScript
Last push
Oct 23, 2026
Topics
codex · cordis · deepseek · dsh · dsh-plugin

Good for

Getting started

  1. Run npx skills add liustack/modlens; on DeepSeek Harness use npx -y @deepseek-ai/dsh plugin --profile web add @liustack/modlens@3.27.2
  2. Reuse your existing Claude Code or Codex setup, or use the free Antigravity CLI or a Gemini key as the vision source
  3. Paste an image into the chat and let modlens read it

Before you install

It only takes over models confirmed text-only (DeepSeek, GLM, MiMo Pro); native vision models such as GLM-5.3-Flash keep their normal paste behavior. Reading images needs a vision-capable source, via a free or OpenAI-compatible provider key.

Summarized by AI from the project docs; check the official docs before relying on it.

Similar tools

More in Community skills →

Source GitHub · All rights belong to the original creators