modlens
Adds visual capabilities to pure text models; paste an image to get structured JSON evidence (OCR, layout, semantics).
- Type
- Agent Skill
- Open source
- Yes
- GitHub Stars
- ★ 1.3k
- Source
- skill-github
- Repository
- github.com/liustack/modlens
Overview
ModLens is the first visual plugin for DeepSeek Harness, adding visual abilities to pure text models like DeepSeek and GLM. Simply paste an image to receive structured JSON evidence including OCR text, layout information, and semantic analysis. The plugin installs easily with no additional configuration required and supports multiple platforms such as Claude Code, Codex, Pi, and OpenCode. Ideal for developers and researchers needing to convert image information into structured data.
Capabilities
- ▪OCR text recognition
- ▪Layout analysis
- ▪Semantic understanding
- ▪Image-to-structured JSON conversion
Use cases
Setup
npx -y @deepseek-ai/dsh plugin --profile web add @liustack/modlens@3.16.5
This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.
FAQ
How to install ModLens?
Install using the npx command: npx -y @deepseek-ai/dsh plugin --profile web add @liustack/modlens@3.16.5
Which platforms does ModLens support?
Supports platforms such as Claude Code, Codex, Pi, and OpenCode.
Is an API Key required?
You can use a free Gemini API Key or any other OpenAI-compatible API Key, or choose not to use an API Key at all.
Related skills
mantis
Provides security review capabilities for AI coding agents, automatically discovering, reproducing, and fixing vulnerabilities.
sast-skills
Transform your AI coding assistant into a SAST scanner to automatically detect vulnerabilities in code.
Lightswind-UI-Library
AI-native CLI-first React component library with MCP Server support.
scientific-agent-skills
Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.
susi_alexa_skill
A skill that enables question-and-answer interactions between Alexa and Susi AI.
claude-context
Provides code search capabilities for Claude Code, turning the entire codebase into context.