# Gemini Vision

> Analyze images using Google Gemini 2.0 Flash with proxy support, retry logic, and mock mode.

- Skill: `xspoonai/gemini-vision` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add xspoonai/gemini-vision`
- Raw SKILL.md: https://api.skillmd.com/api/skills/xspoonai/gemini-vision/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: XSpoonAi (https://skillmd.com/u/xspoonai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/xspoonai/gemini-vision

---


# Gemini Vision Analyzer

A robust SpoonOS skill for analyzing images using Google's Gemini 2.0 Flash model. It is designed for production environments with built-in support for HTTP proxies, automatic retries for rate limits, and a mock mode for testing without API costs.

## Quick Start

```python
from spoon.skills.gemini_vision import vision_analyze

# Analyze an image
with open("image.jpg", "rb") as f:
    image_bytes = f.read()

result = await vision_analyze(
    image=image_bytes, 
    prompt="Describe this image"
)
print(result)
```

## Scripts

| Script | Purpose |
|--------|---------|
| [main.py](scripts/main.py) | Main implementation of the vision analysis logic |

## Environment Variables

| Variable | Required | Description |
|----------|----------|-------------|
| `GEMINI_API_KEY` | Yes | Your Google Gemini API key |
| `HTTPS_PROXY` | No | Proxy URL (e.g., `http://127.0.0.1:7890`) |
| `MOCK_VISION` | No | Set to `true` to enable mock mode |
| `GEMINI_MODEL` | No | Model name (default: `gemini-2.5-flash`) |

## Best Practices

1. **Use Mock Mode for Dev**: Set `MOCK_VISION=true` during development to save API quota.
2. **Handle Proxies**: If you are in a region where Google API is restricted, ensure `HTTPS_PROXY` is set.
3. **Structured Prompts**: The skill automatically appends JSON formatting instructions, so focus your prompt on *what* to analyze.