Skip to content
OpenAgentSkill
Browsing is open · Catalog captured 2026-11-21 · Automatic updates and account services are paused.
Development

Agent Vision Toolkit

给纯文本 LLM agent 装上眼睛:图片问答、OCR、截图分析、视觉定位等一套视觉工具箱 + skill,并可无缝接入 Codex、Claude Code、OpenCode、Pi | Give text-only LLM agents vision: image Q&A, OCR, screenshot understanding, visual grounding, image-to-SVG - a vision toolkit & skill, with drop-in integration for Codex, Claude Code, OpenCode, Pi

Anionex · AI review on file

Installation

Read the source and dependencies before running a command. Copying does not install or execute anything.

npx skills add Anionex/agent-vision-toolkit

About this skill

agent-vision-toolkit

What it thinks is what it sees — give any text-only coding agent eyes: image Q&A, OCR, screenshot understanding, visual grounding, and image-to-SVG, as a vision toolkit plus a skill, with optional drop-in integration for Codex, Claude Code, Pi, Oh My Pi, and OpenCode.

🌐 中文 | English

If your coding agent runs on a text-only model like DeepSeek V4, it can't look at images — screenshots, mockups, diagrams, and error dialogs are all dead ends. This repository gives it eyes in two layers:

  1. The toolkit — four CLIs, plus a skill that teaches your agent when to reach for each one. Works in any agent with a shell.
  2. Seamless integration (optional upgrade) — a transparent local proxy and single-file native extensions, so pasted images and built-in image tools work too, with no tool call and no extra prompting.

All code has been verified in real Codex + DeepSeek sessions, and the same pipeline has been live-verified end-to-end in Claude Code, Pi, Oh My Pi, and OpenCode. Use cases include but are not limited to: image Q&A, screenshot analysis, Computer Use GUI operation, and multi-step image reasoning.

If this project helps you, feel free to star🌟 & follow~ I'll keep sharing more practical tools and tips.

Real-world Effects

Left: multi-round image Q&A with glance. Right: with ground, DeepSeek V4 locates screen elements to play chess autonomously.

*Left: