360 Secure SkillHub
360 安全技能中心

Vision Analyzer

通过 Python 脚本把本地图片（JPG/PNG/GIF/WebP）连同可选提示词提交到 Ollama Cloud 的 Kimi K2.5 多模态模型，返回图片内容的自然语言描述；需先设置 OLLAMA_API_KEY 环境变量，以命令行单次调用方式运行，脚本路径位于宿主技能目录内。

kawummuwe-stack
kawummuwe-stack
热度 42.87
最近更新 2026-05-11 17:41:08
分类：数据依赖：代码执行 / 读写文件 / 第三方 API / 内部调用大模型应用场景：摘要总结

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add vision-analyzer
Metadata
name
vision-analyzer
description
Analyze images using Ollama Cloud's Kimi K2.5 vision capabilities. Use when user wants to describe, understand, or get information about an image. Works with local image files, screenshots, or downloaded images. Supports JPG, PNG, GIF, WebP formats.
metadata
{"author":"clawdia","version":"1.0.1"}

Vision Analyzer

Analyze images using Kimi K2.5 multimodal vision capabilities through Ollama Cloud API.

When to Use

  • User wants to know what's in an image
  • Describing screenshots or photos
  • Understanding UI elements, text in images
  • Analyzing memes, charts, diagrams

Quick Start

python3 ~/.openclaw/workspace/skills/vision-analyzer/scripts/vision_analyze.py <image_path> [prompt]

Examples

Describe an image:

python3 ~/.openclaw/workspace/skills/vision-analyzer/scripts/vision_analyze.py photo.jpg

Ask specific question:

python3 ~/.openclaw/workspace/skills/vision-analyzer/scripts/vision_analyze.py screenshot.png "What UI elements do you see?"

Common Image Locations

  • Downloads: /mnt/chromeos/MyFiles/Downloads/
  • Screenshots: /mnt/chromeos/MyFiles/Downloads/
  • Home directory: ~/

Configuration

Set your Ollama API key as environment variable:

export OLLAMA_API_KEY="your-api-key-here"

Get your API key from ollama.com/settings

API Configuration

The skill uses Ollama Cloud API with Kimi K2.5 model. API key is read from OLLAMA_API_KEY environment variable.

Supported Formats

  • JPG/JPEG
  • PNG
  • GIF
  • WebP

Output

Returns a natural language description of the image content.