Skip to content

AI Service

The AI Service provides multi-provider AI chat capabilities with conversation management, streaming responses, and both CLI and API interfaces. Choose between Pydantic AI or LangChain as your underlying engine.

AI Service Demo

Start Chatting in 30 Seconds

Generate a project with AI service and start chatting immediately:

uvx aegis-stack init my-app --services ai
cd my-app
uv sync && source .venv/bin/activate
my-app ai chat "Hello! Can you help me understand what you can do?"

No API key required with the PUBLIC provider - perfect for testing!

Requires uv installed. See Installation for other options.

What You Get

  • Multi-Provider Support - OpenAI, Anthropic, Google, Groq, Mistral, Cohere, and free public endpoints
  • Conversation Management - In-memory conversation tracking with message history
  • Streaming Responses - Real-time Server-Sent Events (SSE) streaming for interactive UX
  • CLI Interface - Interactive chat sessions with markdown rendering
  • REST API - Complete /ai/* endpoints for integration
  • Health Monitoring - Service health checks and validation

Architecture

graph TB
    subgraph "AI Service"
        AIService[AI Service<br/>Pydantic AI / LangChain]

        CLI[CLI Interface<br/>ai chat, config, providers]
        API[REST API<br/>/ai/chat, /ai/conversations]
        Providers[Providers<br/>OpenAI, Anthropic, Google<br/>Groq, Mistral, Cohere, PUBLIC]
        Conv[Conversations<br/>In-memory storage]
    end

    Backend[Backend Component<br/>FastAPI]

    AIService --> CLI
    AIService --> API
    AIService --> Providers
    AIService --> Conv
    API --> Backend

    style AIService fill:#e8f5e8,stroke:#2e7d32,stroke-width:3px
    style CLI fill:#e1f5fe,stroke:#1976d2,stroke-width:2px
    style API fill:#e1f5fe,stroke:#1976d2,stroke-width:2px
    style Providers fill:#fff3e0,stroke:#f57c00,stroke-width:2px
    style Conv fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    style Backend fill:#e1f5fe,stroke:#1976d2,stroke-width:2px

Quick Start

1. Generate Project with AI Service

aegis init my-app --services ai
cd my-app
uv sync && source .venv/bin/activate

2. Start Chatting (No API Key Required)

The PUBLIC provider is configured by default and requires no setup:

# Interactive chat session
my-app ai chat

# Single message
my-app ai chat "Explain async/await in Python"

# Check service status
my-app ai status

3. API Usage

Start the server and use the REST API:

# Terminal 1: Start server
make serve

# Terminal 2: Send chat request
curl -X POST http://localhost:8000/ai/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Hello from the API!"}'

Configuration

The AI service uses environment variables for configuration. The PUBLIC provider works with zero configuration:

# .env - Default configuration (PUBLIC provider)
AI_ENABLED=true
AI_PROVIDER=public
AI_MODEL=auto

That's it! You can start chatting immediately with no API keys or setup required.

Using Other Providers

Want faster responses, specific models, or advanced features? You can switch to other providers:

# Example: Switch to Groq for production use
AI_PROVIDER=groq
GROQ_API_KEY=your-key-here
AI_MODEL=llama-3.1-8b-instant

Available providers: OpenAI, Anthropic, Google Gemini, Groq, Mistral, Cohere

→ Complete Provider Setup Guide - Detailed setup for all providers

Features

Streaming Responses

CLI automatically streams for supported providers:

# Streaming enabled by default
my-app ai chat "Write a Python function to calculate fibonacci"

# Disable streaming
my-app ai chat "Same question" --no-stream

API streaming uses Server-Sent Events:

import httpx

async with httpx.AsyncClient() as client:
    async with client.stream(
        "GET",
        "http://localhost:8000/ai/chat/stream",
        params={"message": "Hello"},
    ) as response:
        async for line in response.aiter_lines():
            if line.startswith("data: "):
                data = json.loads(line[6:])
                print(data["content"])  # Stream chunks

Health Monitoring

# Check service health
curl http://localhost:8000/ai/health

# Validate configuration (included in ai status output)
my-app ai status

Next Steps:

Was this page helpful?