aidevops 2.52.1

package/.agent/tools/ai-orchestration/autogen.md

@@ -0,0 +1,406 @@
+---
+description: Microsoft AutoGen multi-agent framework - setup, usage, and integration
+mode: subagent
+tools:
+  read: true
+  write: true
+  edit: true
+  bash: true
+  glob: true
+  grep: true
+  webfetch: true
+---
+
+# AutoGen - Agentic AI Framework
+
+<!-- AI-CONTEXT-START -->
+
+## Quick Reference
+
+- **Purpose**: Programming framework for agentic AI with multi-language support
+- **License**: MIT (code) / CC-BY-4.0 (docs)
+- **Setup**: `bash .agent/scripts/autogen-helper.sh setup`
+- **Start**: `~/.aidevops/scripts/start-autogen-studio.sh`
+- **Stop**: `~/.aidevops/scripts/stop-autogen-studio.sh`
+- **Status**: `~/.aidevops/scripts/autogen-status.sh`
+- **URL**: http://localhost:8081 (AutoGen Studio)
+- **Config**: `~/.aidevops/autogen/.env`
+- **Install**: `pip install autogen-agentchat autogen-ext[openai]`
+
+**Key Features**:
+
+- Multi-language support (Python, .NET)
+- MCP server integration
+- Human-in-the-loop workflows
+- AgentChat for rapid prototyping
+- Core API for advanced control
+
+<!-- AI-CONTEXT-END -->
+
+## Overview
+
+AutoGen is a framework from Microsoft for creating multi-agent AI applications that can act autonomously or work alongside humans. It provides both high-level APIs for rapid prototyping and low-level control for production systems.
+
+## Architecture
+
+AutoGen uses a layered design:
+
+- **Core API**: Message passing, event-driven agents, distributed runtime
+- **AgentChat API**: Simpler API for rapid prototyping
+- **Extensions API**: First and third-party extensions
+
+## Installation
+
+### Automated Setup (Recommended)
+
+```bash
+# Run the setup script
+bash .agent/scripts/autogen-helper.sh setup
+
+# Configure API keys
+nano ~/.aidevops/autogen/.env
+
+# Start AutoGen Studio
+~/.aidevops/scripts/start-autogen-studio.sh
+```
+
+### Manual Installation
+
+```bash
+# Create directory and virtual environment
+mkdir -p ~/.aidevops/autogen
+cd ~/.aidevops/autogen
+python3 -m venv venv
+source venv/bin/activate
+
+# Install AutoGen
+pip install autogen-agentchat autogen-ext[openai]
+
+# Install AutoGen Studio
+pip install autogenstudio
+```
+
+### Quick Start
+
+```bash
+# Export your API key
+export OPENAI_API_KEY="sk-..."
+
+# Run AutoGen Studio
+autogenstudio ui --port 8081
+```
+
+## Configuration
+
+### Environment Variables
+
+Create `~/.aidevops/autogen/.env`:
+
+```bash
+# OpenAI Configuration (Required)
+OPENAI_API_KEY=your_openai_api_key_here
+
+# Anthropic Configuration (Optional)
+ANTHROPIC_API_KEY=your_anthropic_key_here
+
+# Azure OpenAI (Optional)
+AZURE_OPENAI_API_KEY=your_azure_key_here
+AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
+
+# Local LLM (Ollama)
+OLLAMA_BASE_URL=http://localhost:11434
+
+# AutoGen Studio Configuration
+AUTOGEN_STUDIO_PORT=8081
+```
+
+## Usage
+
+### Hello World
+
+```python
+import asyncio
+from autogen_agentchat.agents import AssistantAgent
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+
+async def main():
+    model_client = OpenAIChatCompletionClient(model="gpt-4.1")
+    agent = AssistantAgent("assistant", model_client=model_client)
+    print(await agent.run(task="Say 'Hello World!'"))
+    await model_client.close()
+
+asyncio.run(main())
+```
+
+### MCP Server Integration
+
+```python
+import asyncio
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.ui import Console
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+from autogen_ext.tools.mcp import McpWorkbench, StdioServerParams
+
+async def main():
+    model_client = OpenAIChatCompletionClient(model="gpt-4.1")
+    
+    # Connect to MCP server
+    server_params = StdioServerParams(
+        command="npx",
+        args=["@playwright/mcp@latest", "--headless"]
+    )
+    
+    async with McpWorkbench(server_params) as mcp:
+        agent = AssistantAgent(
+            "web_assistant",
+            model_client=model_client,
+            workbench=mcp,
+            model_client_stream=True,
+            max_tool_iterations=10
+        )
+        await Console(agent.run_stream(task="Search for AutoGen documentation"))
+
+asyncio.run(main())
+```
+
+### Multi-Agent Orchestration
+
+```python
+import asyncio
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.tools import AgentTool
+from autogen_agentchat.ui import Console
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+
+async def main():
+    model_client = OpenAIChatCompletionClient(model="gpt-4.1")
+
+    # Create specialist agents
+    math_agent = AssistantAgent(
+        "math_expert",
+        model_client=model_client,
+        system_message="You are a math expert.",
+        description="A math expert assistant.",
+        model_client_stream=True
+    )
+    math_tool = AgentTool(math_agent, return_value_as_last_message=True)
+
+    chemistry_agent = AssistantAgent(
+        "chemistry_expert",
+        model_client=model_client,
+        system_message="You are a chemistry expert.",
+        description="A chemistry expert assistant.",
+        model_client_stream=True
+    )
+    chemistry_tool = AgentTool(chemistry_agent, return_value_as_last_message=True)
+
+    # Create orchestrator agent
+    orchestrator = AssistantAgent(
+        "assistant",
+        system_message="You are a general assistant. Use expert tools when needed.",
+        model_client=model_client,
+        model_client_stream=True,
+        tools=[math_tool, chemistry_tool],
+        max_tool_iterations=10
+    )
+    
+    await Console(orchestrator.run_stream(task="What is the integral of x^2?"))
+    await Console(orchestrator.run_stream(task="What is the molecular weight of water?"))
+
+asyncio.run(main())
+```
+
+### AutoGen Studio (GUI)
+
+```bash
+# Start AutoGen Studio
+autogenstudio ui --port 8081 --appdir ./my-app
+
+# Access at http://localhost:8081
+```
+
+AutoGen Studio provides a no-code GUI for:
+
+- Building multi-agent workflows
+- Testing agent interactions
+- Prototyping without writing code
+
+## Local LLM Support
+
+### Using Ollama
+
+```python
+from autogen_ext.models.ollama import OllamaChatCompletionClient
+
+model_client = OllamaChatCompletionClient(
+    model="llama3.2",
+    base_url="http://localhost:11434"
+)
+```
+
+### Using Azure OpenAI
+
+```python
+from autogen_ext.models.openai import AzureOpenAIChatCompletionClient
+
+model_client = AzureOpenAIChatCompletionClient(
+    model="gpt-4",
+    azure_endpoint="https://your-resource.openai.azure.com/",
+    api_version="2024-02-15-preview"
+)
+```
+
+## .NET Support
+
+AutoGen also supports .NET for cross-language development:
+
+```csharp
+using Microsoft.AutoGen.Contracts;
+using Microsoft.AutoGen.Core;
+
+// Create an agent in .NET
+var agent = new AssistantAgent("assistant", modelClient);
+var result = await agent.RunAsync("Hello from .NET!");
+```
+
+## Git Integration
+
+### Project Structure
+
+```text
+my-autogen-project/
+├── .env                    # gitignored
+├── agents/
+│   ├── researcher.py
+│   └── writer.py
+├── workflows/
+│   └── research_flow.py
+└── config/
+    └── settings.json       # version controlled
+```
+
+### Version Control Best Practices
+
+```bash
+# .gitignore
+.env
+__pycache__/
+*.pyc
+.venv/
+venv/
+*.log
+.autogen/
+```
+
+## Deployment
+
+### Docker
+
+```dockerfile
+FROM python:3.11-slim
+
+WORKDIR /app
+COPY . .
+
+RUN pip install autogen-agentchat autogen-ext[openai]
+
+CMD ["python", "main.py"]
+```
+
+### FastAPI Integration
+
+```python
+from fastapi import FastAPI
+from autogen_agentchat.agents import AssistantAgent
+from autogen_ext.models.openai import OpenAIChatCompletionClient
+
+app = FastAPI()
+
+@app.post("/chat")
+async def chat(message: str):
+    model_client = OpenAIChatCompletionClient(model="gpt-4o-mini")
+    agent = AssistantAgent("assistant", model_client=model_client)
+    result = await agent.run(task=message)
+    await model_client.close()
+    return {"response": str(result)}
+```
+
+## Integration Examples
+
+### With aidevops Workflows
+
+```python
+# DevOps automation with AutoGen
+from autogen_agentchat.agents import AssistantAgent
+from autogen_agentchat.tools import AgentTool
+
+# Create specialized DevOps agents
+code_reviewer = AssistantAgent(
+    "code_reviewer",
+    system_message="You review code for quality and security issues."
+)
+
+deployment_agent = AssistantAgent(
+    "deployment_agent", 
+    system_message="You handle deployment tasks and CI/CD."
+)
+
+# Orchestrate DevOps workflow
+devops_orchestrator = AssistantAgent(
+    "devops_lead",
+    tools=[
+        AgentTool(code_reviewer),
+        AgentTool(deployment_agent)
+    ]
+)
+```
+
+### With Langflow
+
+AutoGen agents can be wrapped as Langflow custom components.
+
+### With CrewAI
+
+Both frameworks can be used together - AutoGen for conversational flows, CrewAI for role-based teams.
+
+## Troubleshooting
+
+### Common Issues
+
+**Import errors**:
+
+```bash
+# Ensure packages are installed
+pip install autogen-agentchat autogen-ext[openai]
+```
+
+**Async errors**:
+
+```python
+# Always use asyncio.run() for async code
+import asyncio
+asyncio.run(main())
+```
+
+**Model client not closing**:
+
+```python
+# Always close model clients
+await model_client.close()
+
+# Or use context manager
+async with model_client:
+    # ... use client
+```
+
+### Migration from v0.2
+
+If upgrading from AutoGen v0.2, see the [Migration Guide](https://microsoft.github.io/autogen/stable/user-guide/agentchat-user-guide/migration-guide.html).
+
+## Resources
+
+- **Documentation**: https://microsoft.github.io/autogen/
+- **GitHub**: https://github.com/microsoft/autogen
+- **Discord**: https://aka.ms/autogen-discord
+- **Blog**: https://devblogs.microsoft.com/autogen/
+- **PyPI**: https://pypi.org/project/autogen-agentchat/