Habilidades azure-ai-projects-py
📦

azure-ai-projects-py

Seguro

Build Azure AI Apps with Foundry SDK

Build AI applications on Microsoft Foundry using the Azure AI Projects Python SDK. This skill provides comprehensive documentation for creating agents, managing threads, configuring tools, and running evaluations.

Soporta: Claude Codex Code(CC)
🥉 75 Bronce
1

Descargar el ZIP de la skill

2

Subir en Claude

Ve a Configuración → Capacidades → Skills → Subir skill

3

Activa y empieza a usar

Pruébalo

Usando "azure-ai-projects-py". Create a basic agent with CodeInterpreterTool

Resultado esperado:

The agent is created successfully with ID 'agent-123abc'. Tools enabled: CodeInterpreterTool. You can now create threads and runs with this agent.

Usando "azure-ai-projects-py". Show me how to list deployments

Resultado esperado:

Available deployments: ['gpt-4o-mini', 'gpt-4o', 'o1-mini']. You can use these model names when creating agents.

Auditoría de seguridad

Seguro
v1 • 2/24/2026

All 74 static findings are false positives. The file is a markdown documentation file (SKILL.md) containing code examples for the Azure AI Projects Python SDK. The 'external_commands' detections are markdown code fences, not shell execution. The 'network' detection is a placeholder URL template in documentation. The 'env_access' detections are documentation examples showing SDK configuration. No actual executable code or malicious patterns exist. This is a legitimate documentation skill for Microsoft's Azure Foundry SDK.

1
Archivos escaneados
301
Líneas analizadas
0
hallazgos
1
Auditorías totales
No se encontraron problemas de seguridad
Auditado por: claude

Puntuación de calidad

38
Arquitectura
100
Mantenibilidad
87
Contenido
50
Comunidad
100
Seguridad
100
Cumplimiento de la especificación

Lo que puedes crear

Build AI Agents in Microsoft Foundry

Create intelligent agents that can execute Python code, search files, and call external APIs using the Azure AI Projects SDK.

Configure Agent Tools and Capabilities

Add capabilities like code interpretation, file search, Bing grounding, and custom function calling to your agents.

Run Evaluations on Agent Outputs

Use built-in evaluators or create custom evaluators to test and measure agent quality and adherence to requirements.

Prueba estos prompts

Create Basic Agent 
Create an Azure AI Projects agent named 'my-agent' using the gpt-4o-mini model with the instruction 'You are a helpful assistant.' Use the AIProjectClient with DefaultAzureCredential.
Agent with Code Interpreter 
Create an agent with CodeInterpreterTool and FileSearchTool enabled. The agent should be able to execute Python code and search through uploaded documents.
Conversation Thread Flow 
Show me how to create a thread, add a user message, create and process a run, and retrieve the assistant response using the Azure AI Projects SDK.
Versioned Agent for Production 
Create a versioned agent using PromptAgentDefinition with version label 'v1.0'. Include instructions for a customer support specialist and show how to deploy it.

Mejores prácticas

  • Use context managers (async with) for the async AIProjectClient to ensure proper resource cleanup
  • Create versioned agents with PromptAgentDefinition for production deployments to maintain stable APIs
  • Clean up agents with client.agents.delete_agent(agent.id) when no longer needed to avoid resource leaks

Evitar

  • Do not hardcode credentials in code - always use DefaultAzureCredential or environment variables
  • Do not use the low-level AgentsClient when you need Foundry-specific features like connections and deployments
  • Avoid creating many agents without cleanup - always delete agents when done

Preguntas frecuentes

What is the difference between AIProjectClient and AgentsClient?
AIProjectClient is a high-level client for full Foundry integration with connections, deployments, datasets, and versioning. AgentsClient is a low-level client for standalone agent applications without Foundry features.
How do I authenticate to Azure AI Projects?
Use DefaultAzureCredential which automatically handles authentication through Azure AD, managed identities, or environment variables.
Can I use OpenAI API directly with this SDK?
Yes, use client.get_openai_client() to get an OpenAI-compatible client for standard OpenAI API calls.
What tools can I add to my agents?
You can add CodeInterpreterTool, FileSearchTool, BingGroundingTool, AzureAISearchTool, FunctionTool, OpenApiTool, McpTool, MemorySearchTool, and SharepointGroundingTool.
How do I create versioned agents?
Use client.agents.create_version() with a PromptAgentDefinition to create versioned agents with stable APIs for production.
Does this support async operations?
Yes, use AIProjectClient from azure.ai.projects.aio for async operations with async/await patterns.

Detalles del desarrollador

Autor

sickn33

Licencia

MIT

Repositorio

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/azure-ai-projects-py

Ref.

main

Estructura de archivos

📄 SKILL.md