AGENT SKILL

openai-vision

Analyze images and multi-frame sequences using OpenAI GPT vision models

License 许可
Apache-2.0
最近更新
2026-11-14
Stars
★ 0
Forks
1

01 前言 Front Matter

name
openai-vision
description
Analyze images and multi-frame sequences using OpenAI GPT vision models

03 SKILL.md 正文

(unchanged — see original skill content)

Steps

  1. Encode local images to base64 with correct media type, or pass URLs directly.
  2. Call client.chat.completions.create with model="gpt-4o" and a content array mixing text and image_url parts.
  3. Choose detail: high for OCR / small objects, low for scene classification, auto otherwise.
  4. For structured output, instruct the model to return only JSON and parse it, stripping code fences.
  5. Handle rate limits with exponential backoff; resize images over ~15MB before sending.

Expected Result

A parsed analysis (string or JSON) plus token-usage metadata for each processed image.

04 同仓库更多 Skills