astra-framework 0.1.0

Astra Framework - Core APIs and abstractions for building AI agents, teams, RAG pipelines, and workflows

Astra Framework

🧠 Core framework for building AI agents, teams, and RAG pipelines.

This is the foundational layer of Astra - providing all the building blocks for intelligent AI applications.

Installation

pip install astra-framework

Optional Dependencies

# With MongoDB support
pip install astra-framework[mongodb]

# With AWS Bedrock support
pip install astra-framework[aws]

# Everything
pip install astra-framework[all]

Features

Component	Description
🤖 Agents	Intelligent agents with tools, memory, and context
📚 RAG	Retrieval-Augmented Generation pipelines
🗄️ Storage	LibSQL, MongoDB backends
🛡️ Guardrails	PII filtering, content moderation, injection detection
🔧 Tools	Function calling with @tool decorator
👥 Teams	Multi-agent collaboration and delegation
💾 Memory	Short-term and long-term agent memory
🔌 Middleware	Input/output processing pipelines

Model Support

• Google Gemini
• OpenAI GPT
• AWS Bedrock (Claude, etc.)
• HuggingFace Local Models

Quick Example

from framework.agents import Agent, tool
from framework.models import Gemini

@tool
def get_weather(city: str) -> str:
    """Get weather for a city."""
    return f"Sunny in {city}"

agent = Agent(
    model=Gemini("gemini-2.0-flash"),
    instructions="You are a helpful assistant",
    tools=[get_weather]
)

response = await agent.invoke("What's the weather in Tokyo?")

Documentation

See examples for comprehensive examples.

Related Packages

• astra-runtime: High-level runtime with embedded mode and FastAPI server
• astra-observability: Tracing, metrics, and logging

License

MIT License