# Analyze Images with Azure AI Vision

Manual image inspection limits scale. This skill guides Azure AI Vision workflows for captions, OCR, objects, people, and smart crops.

## Install

```bash
npx skillstore add sickn33/azure-ai-vision-imageanalysis-py
```

## Metadata

- Status: approved
- Slug: sickn33-azure-ai-vision-imageanalysis-py
- Skillstore revision: r2
- Version status: missing
- Tree hash: f7e8328d05dfd001517c5ee0f94f188c0ef0eed1cea2f835e321928d604af7ef
- Author: sickn33
- GitHub username: sickn33
- License: MIT
- Repository: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/azure-ai-vision-imageanalysis-py
- Ref: f9e2c34b4f19c7f3e6b0a1e93227b5f77cc12526
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, env\_access
- Quality score: 79
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/sickn33-azure-ai-vision-imageanalysis-py
- Manifest: https://skillstore.pages.dev/api/skills/sickn33-azure-ai-vision-imageanalysis-py/manifest

## Capabilities

- Configures Azure AI Vision clients with API keys or Entra ID credentials.
- Analyzes remote image URLs and local image files.
- Requests captions, dense captions, tags, objects, OCR, and people detection.
- Extracts confidence values, bounding boxes, and text polygons from results.
- Requests smart crop regions for selected aspect ratios.
- Demonstrates asynchronous analysis and Azure response error handling.

## Use Cases

- Build Accessible Image Descriptions: Generate concise captions and confidence values for image accessibility workflows.
- Extract Document Text: Read lines, words, confidence values, and polygons from photographed documents.
- Prepare Media Thumbnails: Detect content and request crop regions for portrait, standard, and widescreen thumbnails.

## Prompt Templates

### Create a Basic Caption Example

```
Show a minimal Python example that captions an image URL with Azure AI Vision and prints the caption confidence.
```

### Extract OCR from a File

```
Create a Python workflow that reads [IMAGE_PATH], extracts lines and words, and reports confidence values and bounding polygons.
```

### Combine Visual Features

```
Design an Azure AI Vision request for [IMAGE_URL] using captions, tags, objects, people detection, and smart crops at [ASPECT_RATIOS].
```

### Plan an Async Analysis Pipeline

```
Design an asynchronous Python pipeline for [VOLUME] images with Entra ID, bounded concurrency, retries, caching, and HttpResponseError handling.
```

## Limitations

- Requires an Azure AI Vision resource, credentials, and network access.
- Supports only image formats, dimensions, and file sizes accepted by Azure AI Vision.
- Example outputs depend on service availability, model behavior, language support, and image quality.
- Provides implementation guidance, but does not deploy resources or validate production security settings.

## Best Practices

- Request only required visual features to control latency and service cost.
- Prefer Entra ID and keep credentials outside source code.
- Validate inputs, handle HttpResponseError, and cache repeated analyses when appropriate.

## Anti Patterns

- Do not hardcode Azure credentials in source files or prompts.
- Do not request every visual feature when the workflow uses only one result type.
- Do not treat captions, detections, or OCR text as guaranteed facts without confidence checks.

## Security Audit

- Audited at: 2026-07-23T20:02:50.339\+00:00
- Summary: All 42 static findings are false positives. They match Markdown formatting, Azure placeholders, or documented environment configuration. No malicious intent or prompt injection appears.

## Stats

- Views: 82
- Downloads: 10
- Favorites: 0
- Popularity score: 0
