Skills azure-ai

☁️

azure-ai

Name: azure-ai
Author: microsoft

Safe

Build with Azure AI Services

This skill provides comprehensive documentation and SDK references for Azure AI services including Search, Speech, OpenAI, Vision, Translation, and Document Intelligence. It helps developers quickly find the right APIs and code patterns for their AI solutions.

Supports: Claude Codex Code(CC)

🥉 73 Bronze

Download the skill ZIP

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

Toggle on and start using

Test it

Using "azure-ai". How do I set up hybrid search combining keyword and vector search in Azure AI Search?

Expected outcome:

Use the SearchClient with both search_text and vectorQueries. Configure a semantic configuration for best results. Example: client.search(search_text='query', vector_queries=[VectorizedQuery(vector=embedding, k_nearest=3, fields='embedding')])

Using "azure-ai". Show me how to authenticate with Azure OpenAI using the .NET SDK

Expected outcome:

Use AzureKeyCredential with your endpoint and API key: var client = new OpenAIClient(new Uri(endpoint), new AzureKeyCredential(apiKey)); Then call client.GetChatCompletionsAsync with your deployment name.

Security Audit

Safe

v1 • 2/20/2026

All 195 static findings are false positives. The skill contains documentation with code examples for Azure AI SDK usage. Detected patterns (external_commands, env_access, network) are Markdown formatting artifacts and standard SDK documentation patterns, not security vulnerabilities.

Files scanned

519

Lines analyzed

findings

Total audits

Low Risk Issues (4)

SKILL.md:12-30 references/sdk/azure-ai-contentsafety-py.md:7-32

False Positive: Shell Command Detection in Documentation

164 instances of 'external_commands' flagged backticks in Markdown code blocks. These are documentation examples (pip install, npm install), not executable code.

references/sdk/azure-ai-translation-ts.md:14-15

False Positive: Environment Variable Access in Documentation

8 instances flagged environment variable access (process.env.TRANSLATOR_SUBSCRIPTION_KEY). These are standard Azure SDK credential patterns shown in documentation.

SKILL.md:66-67

False Positive: Hardcoded URLs in Documentation

2 instances flagged hardcoded URLs. These are legitimate Microsoft Learn documentation links.

SKILL.md:3-52 references/sdk/azure-ai-vision-imageanalysis-py.md:27-28

False Positive: Cryptographic Algorithm Detection

4 instances flagged 'weak cryptographic algorithm'. Triggered by words like 'analysis', 'stemming', 'neutral' in documentation text.

Audited by: claude

Quality Score

Architecture

100

Maintainability

Content

Community

Security

Spec Compliance

What You Can Build

Build intelligent search applications

Create search experiences with semantic understanding using vector embeddings and hybrid search capabilities

Add speech capabilities to apps

Integrate real-time transcription and text-to-speech for accessibility and voice interfaces

Process documents at scale

Extract text and data from PDFs, forms, and images using OCR and form recognition

Try These Prompts

Basic search query

How do I use Azure AI Search to perform a vector search in Python?

Speech transcription

Show me how to transcribe audio files to text using Azure Speech service with Python

Document extraction

What is the code to extract text from a PDF using Azure Document Intelligence in TypeScript?

Image analysis

Give me an example of using Azure Vision Image Analysis to get captions and tags from an image URL

Best Practices

Use hybrid search combining keyword and vector queries for best relevance
Enable semantic ranking for natural language query understanding
Process documents in batches of 100-1000 for optimal performance
Use async clients for high-throughput scenarios

Avoid

Do not hardcode credentials - use environment variables or Azure Managed Identity
Avoid querying large datasets without pagination or batching
Do not skip error handling for API calls
Avoid using deprecated SDK versions - keep packages updated

Frequently Asked Questions

What is the difference between AI Search and Azure OpenAI?

AI Search provides search and indexing capabilities with vector support. Azure OpenAI provides access to GPT models for generative AI. They can be used together for RAG applications.

Do I need an Azure subscription to use this skill?

Yes, Azure AI services require an Azure subscription. You can use free tiers for development, but production usage requires paid subscriptions.

Can I use this skill offline?

No, Azure AI services are cloud-based and require network connectivity to access the APIs.

Which programming languages are supported?

Azure AI services support Python, TypeScript, Java, .NET, and Go. This skill includes quick reference guides for Python, TypeScript, .NET, and Java.

How do I enable the MCP server for Azure AI?

Run /azure:setup or enable via /mcp command. The MCP server provides tool-based access to AI Search and Speech services.

What are the rate limits for Azure AI services?

Rate limits vary by service and pricing tier. Check Azure portal for your specific limits. Use exponential backoff for retries.