Skip to content

Commit 4a9c934

Browse files
mrunalhirve128mrunalhirve128
committed
fix
1 parent a8a8d1e commit 4a9c934

2 files changed

Lines changed: 24 additions & 258 deletions

File tree

  • libraries
    • microsoft-agents-a365-tooling-extensions-azureaifoundry
    • microsoft-agents-a365-tooling-extensions-openai

‎libraries/microsoft-agents-a365-tooling-extensions-azureaifoundry/README.md‎

Lines changed: 0 additions & 127 deletions
Original file line numberDiff line numberDiff line change
@@ -59,130 +59,3 @@ For issues, questions, or feedback:
5959
Copyright (c) Microsoft Corporation. All rights reserved.
6060

6161
Licensed under the MIT License - see the [LICENSE](../../../LICENSE.md) file for details.
62-
63-
credential = DefaultAzureCredential()
64-
project_client = AIProjectClient(
65-
credential=credential,
66-
endpoint="your-project-endpoint",
67-
subscription_id="your-subscription-id",
68-
resource_group_name="your-resource-group",
69-
project_name="your-project-name"
70-
)
71-
72-
# Create MCP tool registration service
73-
registration_service = McpToolRegistrationService()
74-
75-
# Add MCP tool servers to your Azure AI Foundry agent
76-
await registration_service.add_tool_servers_to_agent(
77-
project_client=project_client,
78-
agent_id="your-agent-id",
79-
environment_id="prod",
80-
auth_token="your-auth-token"
81-
)
82-
```
83-
84-
## Core Azure AI Foundry Components
85-
86-
| Component | Purpose | Description |
87-
|-----------|---------|-------------|
88-
| **McpToolRegistrationService** | Tool Registration | Register MCP servers with Azure AI Foundry agents |
89-
| **Azure Identity Integration** | Authentication | Seamless Azure authentication and credential management |
90-
| **Tool Resource Management** | Resource Management | Manage tool definitions and resources in Azure AI Foundry |
91-
| **Project Client Integration** | Azure Integration | Native integration with Azure AI Projects and clients |
92-
93-
## Supported Azure AI Foundry Features
94-
95-
| Feature | Component | Description |
96-
|---------|-----------|-------------|
97-
| **Agent Tool Registration** | MCP server integration | Add MCP tool servers to Azure AI Foundry agents |
98-
| **Azure Authentication** | Identity management | Azure DefaultAzureCredential integration |
99-
| **Project Integration** | Azure AI Projects | Native integration with Azure AI project clients |
100-
| **Resource Management** | Tool resources | Manage tool definitions and configurations |
101-
102-
## Advanced Usage
103-
104-
### Azure AI Foundry Integration
105-
106-
- **Agent Configuration**: Register MCP tools with existing Azure AI Foundry agents
107-
- **Project Management**: Integration with Azure AI Projects for centralized management
108-
- **Resource Allocation**: Automatic tool resource allocation and configuration
109-
- **Environment Management**: Support for multiple deployment environments
110-
111-
### MCP Tool Registration
112-
113-
The service provides comprehensive MCP tool registration capabilities:
114-
- Automatic discovery of available MCP tool servers
115-
- Registration with Azure AI Foundry agent infrastructure
116-
- Tool resource management and configuration
117-
- Support for both development and production scenarios
118-
119-
## Architecture
120-
121-
The Azure AI Foundry extensions follow a service-oriented architecture:
122-
123-
- **Registration Layer**: MCP tool registration and management services
124-
- **Azure Integration Layer**: Native Azure AI Foundry and identity integration
125-
- **Resource Management Layer**: Tool definition and resource management
126-
- **Configuration Layer**: Environment-specific configuration and settings
127-
128-
## Integration with Microsoft Agent 365 SDK
129-
130-
This package works seamlessly with other Microsoft Agent 365 SDK components:
131-
132-
| Package | Integration |
133-
|---------|-------------|
134-
| `microsoft-agents-a365-tooling` | Core tooling utilities and MCP server configuration |
135-
| `microsoft-agents-a365-runtime` | Runtime utilities and environment management |
136-
| `microsoft-agents-a365-observability-core` | Observability and monitoring for Azure deployments |
137-
| `microsoft-agents-a365-hosting-core` | Agent hosting and middleware services |
138-
139-
## Sample Applications
140-
141-
Check out these working examples:
142-
143-
| Sample | Description | Location |
144-
|--------|-------------|----------|
145-
| **Azure AI Foundry Agent** | Basic agent with MCP tool integration | `samples/azure-foundry-agent/` |
146-
| **Multi-Tool Agent** | Agent with multiple MCP tool servers | `samples/multi-tool-azure-agent/` |
147-
| **Enterprise Azure Deployment** | Production-ready Azure AI Foundry deployment | `samples/enterprise-azure-deployment/` |
148-
149-
## Requirements
150-
151-
- **Python**: 3.11+
152-
- **Dependencies**:
153-
- `microsoft-agents-a365-tooling >= 0.1.0`
154-
- `azure-ai-projects >= 1.0.0`
155-
- `azure-ai-agents >= 1.1.0b4`
156-
- `azure-identity >= 1.12.0`
157-
158-
## Common Use Cases
159-
160-
### Azure AI Foundry Development
161-
- Register MCP tool servers with Azure AI Foundry agents
162-
- Integrate with Azure AI Projects for centralized agent management
163-
- Leverage Azure identity for secure authentication and authorization
164-
- Deploy agents with comprehensive tooling capabilities
165-
166-
### Enterprise Azure Deployment
167-
- Manage MCP tool registration across multiple Azure environments
168-
- Integrate with existing Azure infrastructure and identity systems
169-
- Support multi-tenant deployments with Azure AI Foundry
170-
- Enable enterprise-scale agent deployments with Azure tooling
171-
172-
### Development and Testing
173-
- Test MCP tool integration in Azure AI Foundry environments
174-
- Validate agent configurations with Azure AI Projects
175-
- Support development workflows with Azure tooling integration
176-
- Enable debugging and monitoring with Azure observability
177-
178-
## Quick Links
179-
180-
📦 [All SDK Packages on PyPI](TODO: Update when packages are published)
181-
📖 [Complete Documentation](https://github.com/microsoft/Agent365/tree/main/python)
182-
💡 [Python Samples Repository](https://github.com/microsoft/Agent365/tree/main/samples)
183-
🐛 [Report Issues](https://github.com/microsoft/Agent365/issues)
184-
☁️ [Azure AI Foundry Documentation](https://learn.microsoft.com/azure/ai-studio/)
185-
186-
## License
187-
188-
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

‎libraries/microsoft-agents-a365-tooling-extensions-openai/README.md‎

Lines changed: 24 additions & 131 deletions
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
[![PyPI](https://img.shields.io/pypi/v/microsoft-agents-a365-tooling-extensions-openai?label=PyPI&logo=pypi)](https://pypi.org/project/microsoft-agents-a365-tooling-extensions-openai)
44
[![PyPI Downloads](https://img.shields.io/pypi/dm/microsoft-agents-a365-tooling-extensions-openai?label=Downloads&logo=pypi)](https://pypi.org/project/microsoft-agents-a365-tooling-extensions-openai)
55

6-
OpenAI Agents SDK specific tools and services for AI agent development. Provides sample implementations and utilities for integrating MCP (Model Context Protocol) servers with OpenAI-based agents.
6+
OpenAI Agents SDK specific tools and services for AI agent development. Provides MCP (Model Context Protocol) tool registration service for dynamically adding MCP servers to OpenAI Agents SDK-based agents.
77

88
## Installation
99

@@ -13,23 +13,34 @@ pip install microsoft-agents-a365-tooling-extensions-openai
1313

1414
## Usage
1515

16-
This package provides sample implementations and patterns for OpenAI agent integration. Refer to the included sample files for implementation examples:
16+
### Basic MCP Tool Registration
1717

18-
- `mcp_demo.py`: Complete working demonstration
19-
- `sample_agent.py`: Basic agent with MCP management
20-
- `advanced_tool_service.py`: Advanced service with multiple server types
21-
- `tool_service_interface.py`: Interface definitions and patterns
18+
```python
19+
from microsoft_agents_a365.tooling.extensions.openai import (
20+
McpToolRegistrationService
21+
)
22+
from agents import Agent
2223

23-
### Basic Pattern
24+
# Initialize the tool registration service
25+
mcp_service = McpToolRegistrationService()
2426

25-
```python
26-
from openai import OpenAI
27+
# Create your OpenAI agent
28+
agent = Agent(
29+
name="my-agent",
30+
instructions="You are a helpful assistant"
31+
)
2732

28-
# Initialize OpenAI client
29-
client = OpenAI(api_key="your-api-key")
33+
# Add MCP tool servers to agent
34+
await mcp_service.add_tool_servers_to_agent(
35+
agent=agent,
36+
agentic_app_id="your-agent-id",
37+
environment_id="your-environment-id",
38+
auth=authorization,
39+
context=turn_context
40+
)
3041

31-
# Use MCP tool discovery patterns from samples
32-
# See package samples for complete implementation examples
42+
# Use the agent with registered MCP tools
43+
response = await agent.run("Help me with my task")
3344
```
3445

3546
## Support
@@ -44,121 +55,3 @@ For issues, questions, or feedback:
4455
Copyright (c) Microsoft Corporation. All rights reserved.
4556

4657
Licensed under the MIT License - see the [LICENSE](../../../LICENSE.md) file for details.
47-
48-
registration_service = McpToolRegistrationService()
49-
50-
# Add MCP tool servers to your OpenAI agent
51-
await registration_service.add_tool_servers_to_agent(
52-
agent=your_openai_agent,
53-
agentic_app_id="user-123",
54-
environment_id="prod",
55-
auth=authorization_context,
56-
context=turn_context,
57-
auth_token="your-auth-token"
58-
)
59-
```
60-
61-
## Core OpenAI Integration Components
62-
63-
| Component | Purpose | Description |
64-
|-----------|---------|-------------|
65-
| **McpToolRegistrationService** | Tool Registration | Register MCP servers with OpenAI Agents SDK applications |
66-
| **MCPServerInfo** | Server Configuration | Configuration model for MCP server information and settings |
67-
| **Multiple Server Types** | Server Support | Support for hosted, streamable HTTP, SSE, and stdio servers |
68-
| **Authentication Integration** | Security | Integration with Microsoft Agent 365 authentication systems |
69-
70-
## Supported MCP Server Types
71-
72-
| Server Type | Component | Description |
73-
|-------------|-----------|-------------|
74-
| **Hosted** | MCPServerHosted | Fully hosted MCP servers with remote endpoints |
75-
| **Streamable HTTP** | MCPServerStreamableHttp | HTTP-based streaming MCP servers |
76-
| **Server-Sent Events** | MCPServerSSE | SSE-based real-time MCP servers |
77-
| **Standard I/O** | MCPServerStdio | Process-based MCP servers using stdio communication |
78-
79-
## Advanced Usage
80-
81-
### OpenAI Agents Integration
82-
83-
- **Agent Configuration**: Register MCP tools with existing OpenAI Agents SDK applications
84-
- **Dynamic Tool Addition**: Runtime addition of MCP tool servers to active agents
85-
- **Server Type Selection**: Choose appropriate MCP server types for different scenarios
86-
- **Authentication Flow**: Seamless integration with Microsoft Agent 365 authentication
87-
88-
### MCP Tool Management
89-
90-
The service provides comprehensive MCP tool management capabilities:
91-
- Automatic discovery of available MCP tool servers
92-
- Registration with OpenAI Agents SDK framework
93-
- Support for multiple concurrent MCP server types
94-
- Runtime management of tool server configurations
95-
96-
## Architecture
97-
98-
The OpenAI tooling extensions follow a service-oriented architecture:
99-
100-
- **Registration Layer**: MCP tool registration and management services
101-
- **OpenAI Integration Layer**: Native OpenAI Agents SDK integration and compatibility
102-
- **Server Management Layer**: Multi-type MCP server support and configuration
103-
- **Authentication Layer**: Microsoft Agent 365 authentication and authorization
104-
105-
## Integration with Microsoft Agent 365 SDK
106-
107-
This package works seamlessly with other Microsoft Agent 365 SDK components:
108-
109-
| Package | Integration |
110-
|---------|-------------|
111-
| `microsoft-agents-a365-tooling` | Core tooling utilities and MCP server configuration |
112-
| `microsoft-agents-a365-runtime` | Runtime utilities and environment management |
113-
| `microsoft-agents-a365-hosting-core` | Agent hosting and middleware services |
114-
| `microsoft-agents-a365-observability-extensions-openai` | OpenAI observability and monitoring |
115-
116-
## Sample Applications
117-
118-
Check out these working examples:
119-
120-
| Sample | Description | Location |
121-
|--------|-------------|----------|
122-
| **OpenAI MCP Agent** | Basic OpenAI agent with MCP tool integration | `samples/openai-mcp-agent/` |
123-
| **Multi-Tool OpenAI Agent** | Agent with multiple MCP server types | `samples/multi-tool-openai-agent/` |
124-
| **Enterprise OpenAI Deployment** | Production-ready OpenAI agent deployment | `samples/enterprise-openai-deployment/` |
125-
126-
## Requirements
127-
128-
- **Python**: 3.11+
129-
- **Dependencies**:
130-
- `microsoft-agents-a365-tooling >= 0.1.0`
131-
- `openai-agents`
132-
- `asyncio-throttle`
133-
134-
## Common Use Cases
135-
136-
### OpenAI Agents Development
137-
- Register MCP tool servers with OpenAI Agents SDK applications
138-
- Integrate with existing OpenAI agent workflows and frameworks
139-
- Leverage Microsoft Agent 365 tooling infrastructure with OpenAI agents
140-
- Deploy agents with comprehensive MCP tooling capabilities
141-
142-
### Enterprise OpenAI Deployment
143-
- Manage MCP tool registration across multiple OpenAI environments
144-
- Integrate with existing Microsoft Agent 365 infrastructure
145-
- Support multi-tenant deployments with OpenAI Agents SDK
146-
- Enable enterprise-scale agent deployments with OpenAI tooling
147-
148-
### Development and Testing
149-
- Test MCP tool integration with OpenAI Agents SDK
150-
- Validate agent configurations with various MCP server types
151-
- Support development workflows with OpenAI tooling integration
152-
- Enable debugging and monitoring with OpenAI observability
153-
154-
## Quick Links
155-
156-
📦 [All SDK Packages on PyPI](TODO: Update when packages are published)
157-
📖 [Complete Documentation](https://github.com/microsoft/Agent365/tree/main/python)
158-
💡 [Python Samples Repository](https://github.com/microsoft/Agent365/tree/main/samples)
159-
🐛 [Report Issues](https://github.com/microsoft/Agent365/issues)
160-
🤖 [OpenAI Agents Documentation](https://platform.openai.com/docs/assistants/overview)
161-
162-
## License
163-
164-
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

0 commit comments

Comments
 (0)