AgentMatter
EN/中文
SubmitGitHub
DSH Plugins · Publisher repository

ModLens

Adds image understanding, OCR, layout analysis, and semantic evidence to text-only coding agents.

1.8k StarsMITTypeScriptUpdated 3 days ago
Install and use GitHub
01

Project overview

Adds image understanding, OCR, layout analysis, and semantic evidence to text-only coding agents. Its core areas include OCR, Image understanding, Layout analysis. Start with the repository documentation and the smallest practical scope before adopting it broadly.

02

Core capabilities

01

OCR

Adds a repeatable workflow or integration for OCR.

02

Image understanding

Adds a repeatable workflow or integration for Image understanding.

03

Layout analysis

Adds a repeatable workflow or integration for Layout analysis.

03

Installation and usage

This usage path is based on the current public documentation for ModLens. Complete the steps in order before using it from your agent.

AI AGENT INSTALL

Let an AI Agent install it

Send this prompt to Codex, Claude Code, or another AI agent that can work with your local environment.

Help me install this DSH: ModLens
Project URL: https://github.com/liustack/modlens

Read the README and installation files first, confirm the current environment and target directory, then follow the project's documented installation method. Run a minimal verification task afterward and report the install location, steps, and result. Ask before requesting credentials, additional permissions, overwriting files, or performing risky actions.
01Before you start
  • A vision model or compatible provider
02Copy the install command or configuration
Project URL: https://github.com/liustack/modlens/blob/main/INSTALL.md
Copy the URL and follow the steps below to complete setup.
03Complete the setup steps
  1. 1
    Apply the configuration

    Use the command or configuration above to add the resource to the current agent environment.

  2. 2
    Complete required setup

    Prepare A vision model or compatible provider.

  3. 3
    Run a minimal task

    Run a minimal task in the agent. If the resource is recognized and returns the documented type of result, the setup is active.

How to verify the setup

Run a minimal task in the agent. If the resource is recognized and returns the documented type of result, the setup is active.

04

Use cases

SCENARIO 01

You use DeepSeek Harness and want additional tools, interface features, or engineering workflows

SCENARIO 02

Your current task needs OCR or Image understanding

05

Assessment

Why it may be useful

  • You use DeepSeek Harness and want additional tools, interface features, or engineering workflows
  • Your current task needs OCR or Image understanding

What to know first

  • You do not have a DSH environment and the project does not document another host
  • You want to use it in production without reading the repository documentation or reviewing permissions
06

README

ModLens

Overview

Adds image understanding, OCR, layout analysis, and semantic evidence to text-only coding agents. The first vision plugin for DeepSeek Harness, and a vision bridge for text-only coding agents.

Getting started

  • This usage path is based on the current public documentation for ModLens. Complete the steps in order before using it from your agent.
  • Project focus: Adds image understanding, OCR, layout analysis, and semantic evidence to text-only coding agents.
  • Published characteristics: 5 vision providers, Structured JSON, Failover.
  • Run a minimal task in the agent. If the resource is recognized and returns the documented type of result, the setup is active.

Configuration

Project URL: https://github.com/liustack/modlens/blob/main/INSTALL.md
Copy the URL and follow the steps below to complete setup.
Read the complete README on GitHub →