# codex-vision-bridge

> codex-vision-bridge — zouyuanqing-codex-vision-bridge. Use this tool when you need to enhance text-only large language models (LLMs) with visual capabilities, such as describing images, locating objects, or performing OCR. The codex-vision-bridge provides an interactive interface for vision primitives, accepting image inputs and generating text outputs with coordinates, annotations, and scanned anomalies. Ideal for use cases requiring automated image analysis and understanding, such as data annotation, quality control, or content moderation.

Canonical page: https://skillsregistry.net/skills/zouyuanqing-codex-vision-bridge  
JSON: https://api.skillsregistry.net/v1/skills/zouyuanqing-codex-vision-bridge

## Description

Interactive vision-primitive MCP server for text-only LLMs (Codex/DeepSeek): describe, locate (coordinates), OCR with bbox, annotate, crop, zoom, and automated anomaly scanning — powered by Xiaomi MiMo V2.5. Zero-dependency single-file Python.

## Trust

- **Trust score (0–1):** 0.78
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **License:** MIT
- **Updated:** 2026-09-28

## Source

- **Source listing:** [GitHub](https://github.com/zouyuanqing/codex-vision-bridge)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "zouyuanqing-codex-vision-bridge"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/zouyuanqing-codex-vision-bridge` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/zouyuanqing-codex-vision-bridge/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt