SkillAgentSearch skills...

openai-vision

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

Install / Use

npx skills add benchflow-ai/skillsbench --skill openai-vision

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

86/100

Supported Platforms

Universal

Our assessment of openai-vision

openai-vision scores 86/100 on our quality scale, 541st of 947 AI & Machine Learning skills we index.

Its SKILL.md is 20 KB long, well organised into 30 sections with 15 code examples: a thorough specification that gives an agent plenty to work with.

With 1,813 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
20/20
Description
12/15
Adoption
14/20
Freshness
11/15

Maintenance, license and trust

  • The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
  • Our last check on 2026-11-02 found the source still online.
  • It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 98/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-10-01. Automated pattern scan on 2026-09-30. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

openai-vision compared with similar skills

All 4 of these similar skills score higher than openai-vision; compare them before choosing.

SkillScoreStarsUpdatedFormat
openai-vision (this skill)by benchflow-ai861.8k4mo agoSKILL.md
claude-memby thedotmack100117.6ktodayCLAUDE.md
taste-skillby Leonxlnx100116.8k8d agoCLAUDE.md
Understand-Anythingby Egonex-AI10092.8k1d agoCLAUDE.md
headroomby headroomlabs-ai10079.8ktodayCLAUDE.md

Frequently asked questions

How do I install openai-vision?
Run npx skills add benchflow-ai/skillsbench --skill openai-vision. The install tabs above show the steps for each supported agent.
Which AI agents does openai-vision work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is openai-vision safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 98/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is openai-vision still maintained?
The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.

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

OpenAI Vision Analysis Skill

Purpose

This skill enables image analysis, scene understanding, text extraction, and multi-frame comparison using OpenAI's vision-capable GPT models (e.g., gpt-4o, gpt-4o-mini). It supports single images, multiple images for comparison, and sequential frames for temporal analysis.

When to Use

  • Analyzing image content (objects, scenes, colors, spatial relationships)
  • Extracting and reading text from images (OCR via vision models)
  • Comparing multiple images to detect differences or changes
  • Processing video frames to understand temporal progression
  • Generating detailed image descriptions or captions
  • Answering questions about visual content

Required Libraries

The following Python libraries are required:

from openai import OpenAI
import base64
import json
import os
from pathlib import Path

Input Requirements

  • File formats: JPG, JPEG, PNG, WEBP, non-animated GIF
  • Image sources: URL, Base64-encoded data, or local file paths
  • Size limits: Up to 20MB per image; total request payload under 50MB
  • Maximum images: Up to 500 images per request
  • Image quality: Clear, legible content; avoid watermarks or heavy distortions

Output Schema

Analysis results should be returned as valid JSON conforming to this schema:

{
  "success": true,
  "images_analyzed": 1,
  "analysis": {
    "description": "A detailed scene description...",
    "objects": [
      {"name": "car", "color": "red", "position": "foreground center"},
      {"name": "tree", "count": 3, "position": "background"}
    ],
    "text_content": "Any text visible in the image...",
    "colors": ["blue", "green", "white"],
    "scene_type": "outdoor/urban"
  },
  "comparison": {
    "differences": ["Object X appeared", "Color changed from A to B"],
    "similarities": ["Background unchanged", "Layout consistent"]
  },
  "metadata": {
    "model_used": "gpt-4o",
    "detail_level": "high",
    "token_usage": {"prompt": 1500, "completion": 200}
  },
  "warnings": []
}

Field Descriptions

  • success: Boolean indicating whether analysis completed
  • images_analyzed: Number of images processed in the request
  • analysis.description: Natural language description of the image content
  • analysis.objects: Array of detected objects with attributes
  • analysis.text_content: Any text extracted from the image
  • analysis.colors: Dominant colors identified
  • analysis.scene_type: Classification of the scene
  • comparison: Present when multiple images are analyzed; describes differences and similarities
  • metadata.model_used: The GPT model used for analysis
  • metadata.detail_level: Resolution level used (low, high, or auto)
  • metadata.token_usage: Token consumption for cost tracking
  • warnings: Array of any issues or limitations encountered

Code Examples

Basic Image Analysis from URL

from openai import OpenAI

client = OpenAI()

def analyze_image_url(image_url, prompt="Describe this image in detail."):
    """Analyze an image from a URL using GPT-4o vision."""
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": prompt},
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": image_url,
                            "detail": "high"
                        }
                    }
                ]
            }
        ],
        max_tokens=1000
    )
    return response.choices[0].message.content

Image Analysis from Local File (Base64)

from openai import OpenAI
import base64

client = OpenAI()

def encode_image_to_base64(image_path):
    """Encode a local image file to base64."""
    with open(image_path, "rb") as image_file:
        return base64.standard_b64encode(image_file.read()).decode("utf-8")

def get_image_media_type(image_path):
    """Determine the media type based on file extension."""
    ext = image_path.lower().split('.')[-1]
    media_types = {
        'jpg': 'image/jpeg',
        'jpeg': 'image/jpeg',
        'png': 'image/png',
        'gif': 'image/gif',
        'webp': 'image/webp'
    }
    return media_types.get(ext, 'image/jpeg')

def analyze_local_image(image_path, prompt="Describe this image in detail."):
    """Analyze a local image file using GPT-4o vision."""
    base64_image = encode_image_to_base64(image_path)
    media_type = get_image_media_type(image_path)
    
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": prompt},
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:{media_type};base64,{base64_image}",
                            "detail": "high"
                        }
                    }
                ]
            }
        ],
        max_tokens=1000
    )
    return response.choices[0].message.content

Multi-Image Comparison

from openai import OpenAI
import base64

client = OpenAI()

def compare_images(image_paths, comparison_prompt=None):
    """Compare multiple images and identify differences."""
    if comparison_prompt is None:
        comparison_prompt = (
            "Compare these images carefully. "
            "List all differences and similarities you observe. "
            "Describe any changes in objects, colors, positions, or text."
        )
    
    content = [{"type": "text", "text": comparison_prompt}]
    
    for i, image_path in enumerate(image_paths):
        base64_image = encode_image_to_base64(image_path)
        media_type = get_image_media_type(image_path)
        
        # Add label for each image
        content.append({
            "type": "text", 
            "text": f"Image {i + 1}:"
        })
        content.append({
            "type": "image_url",
            "image_url": {
                "url": f"data:{media_type};base64,{base64_image}",
                "detail": "high"
            }
        })
    
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": content}],
        max_tokens=2000
    )
    return response.choices[0].message.content

Multi-Frame Video Analysis

from openai import OpenAI
import base64
from pathlib import Path

client = OpenAI()

def analyze_video_frames(frame_paths, analysis_prompt=None):
    """Analyze a sequence of video frames for temporal understanding."""
    if analysis_prompt is None:
        analysis_prompt = (
            "These are sequential frames from a video. "
            "Describe what is happening over time. "
            "Identify any motion, changes, or events that occur across the frames."
        )
    
    content = [{"type": "text", "text": analysis_prompt}]
    
    for i, frame_path in enumerate(frame_paths):
        base64_image = encode_image_to_base64(frame_path)
        media_type = get_image_media_type(frame_path)
        
        content.append({
            "type": "text",
            "text": f"Frame {i + 1}:"
        })
        content.append({
            "type": "image_url",
            "image_url": {
                "url": f"data:{media_type};base64,{base64_image}",
                "detail": "auto"  # Use auto for frames to balance cost
            }
        })
    
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": content}],
        max_tokens=2000
    )
    return response.choices[0].message.content

Full Analysis with JSON Output

from openai import OpenAI
import base64
import json
import os

client = OpenAI()

def analyze_image_to_json(image_path, extract_text=True):
    """Perform comprehensive image analysis and return structured JSON."""
    filename = os.path.basename(image_path)
    
    prompt = """Analyze this image and return a JSON object with the following structure:
{
    "description": "detailed scene description",
    "objects": [{"name": "object name", "attributes": "color, size, position"}],
    "text_content": "any visible text or null if none",
    "colors": ["dominant", "colors"],
    "scene_type": "indoor/outdoor/abstract/etc",
    "people_count": 0,
    "notable_features": ["list of notable visual elements"]
}

Return ONLY valid JSON, no other text."""

    try:
        base64_image = encode_image_to_base64(image_path)
        media_type = get_image_media_type(image_path)
        
        response = client.chat.completions.create(
            model="gpt-4o",
            messages=[
                {
                    "role": "user",
                    "content": [
                        {"type": "text", "text": prompt},
                        {
                            "type": "image_url",
                            "image_url": {
                                "url": f"data:{media_type};base64,{base64_image}",
                                "detail": "high"
                            }
                        }
                    ]
                }
            ],
            max_tokens=1500
        )
        
        # Parse the response as JSON
        analysis_text = response.choices[0].message.content
        # Remove markdown code blocks if present
        if analysis_text.startswith("```"):
            analysis_text = analysis_text.split("```")[1]
            if analysis_text.startswith("json"):
                analysis_text = analysis_text[4:]
        analysis = json.loads(analysis_text.strip())
        
        result = {
            "success": True,
            "filename": filename,
            "analysis": analysis,
            "metadata": {
                "model_used": "gpt-4o",
                "detail_level": "high",
                "token_usage": {
                    "prompt": response.usage.prompt_tokens,
                    "completion": response.usage.completion_tokens
                }
            },
            "warnings": []
        }
        
    except json.JSONDecodeError as e:
        result = {
            "success": False,
            "filename": filename,
            "analysis": {"raw_response": response.choices[0].message.content},
            "metadata": {"model_used": "gpt-4o"},
            "warnings": [f"Failed to parse JSON: {str(e)}"]
        }
    except Exception as e:
        result = {
            "success": False,
            "filename": filename,
            "analysis": {},
            "metadata": {},
            "warnings": [f"Analysis failed: {str(e)}"]
        }
    
    return result

# Usage
result = analyze_image_to_json("photo.jpg")
print(json.dumps(result, indent=2))

Batch Processing Directory

from openai import OpenAI
import base64
import json
from pathlib import Path

client = OpenAI()

def process_image_directory(directory_path, output_file, prompt=None):
    """Process all images in a directory and save results."""
    if prompt is None:
        prompt = "Describe this image briefly, including any visible text."
    
    image_extensions = {'.jpg', '.jpeg', '.png', '.webp', '.gif'}
    results = []
    
    for file_path in sorted(Path(directory_path).iterdir()):
        if file_path.suffix.lower() in image_extensions:
            print(f"Processing: {file_path.name}")
            
            try:
                analysis = analyze_local_image(str(file_path), prompt)
                results.append({
                    "filename": file_path.name,
                    "success": True,
                    "analysis": analysis
                

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars1.8k
CategoryAI
Updated3mo ago
Forks368

Languages

PDDL

Trust signals

98/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 info
openai-vision — Universal Skill: Install & Safety Check | SkillAgent