Metadata-Version: 2.1
Name: medhaai
Version: 0.4.0
Summary: A flexible and powerful multi-agent AI framework for building advanced AI applications
Home-page: https://github.com/sarath59/medhaAI
Author: Sarath Kavuru
Author-email: sarathc225@gmail.com
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: pydantic>=2.0.0
Requires-Dist: pydantic-settings>=2.0.0
Requires-Dist: openai>=1.0.0
Requires-Dist: anthropic>=0.3.0
Requires-Dist: aiohttp>=3.7.4
Requires-Dist: beautifulsoup4>=4.9.3
Requires-Dist: trafilatura>=1.4.0
Requires-Dist: opentelemetry-api>=1.12.0
Requires-Dist: opentelemetry-sdk>=1.12.0
Requires-Dist: asyncio>=3.4.3
Provides-Extra: dev
Requires-Dist: pytest>=6.2.5; extra == "dev"
Requires-Dist: pytest-asyncio>=0.15.1; extra == "dev"
Requires-Dist: black>=21.9b0; extra == "dev"
Requires-Dist: isort>=5.9.3; extra == "dev"

# MedhaAI

MedhaAI is a flexible and powerful multi-agent AI framework for building advanced AI applications. It provides a set of tools and abstractions that allow developers to create complex, collaborative AI systems with ease.

## Features

- Multi-agent system with prioritized agents
- Support for multiple LLM providers (OpenAI, Anthropic)
- Flexible memory systems
- Advanced planning and execution modules
- Customizable toolkit system with advanced web scraping capabilities
- Live streaming of agent activities
- Performance monitoring and tracing
- Asynchronous operations for improved performance

## Installation

To install MedhaAI, simply use pip:

```bash
pip install medhaai
```

## Quick Start

Here's a simple example to get you started with MedhaAI:

```python
import asyncio
from medhaai import MedhaConfig, BaseLLM, BaseAgent, MedhaMultiAgentSystem

async def main():
    config = MedhaConfig()
    llm = BaseLLM.create("openai", api_key=config.openai_api_key)
    agent = BaseAgent.create("general", name="GeneralAgent", role="assistant", llm=llm)
    mas = MedhaMultiAgentSystem(task_decomposition_llm=llm)
    mas.add_agent(agent)
    result = await mas.run_task("Research and summarize the latest advancements in quantum computing")
    print(result)

asyncio.run(main())
```

## Documentation

For more detailed information and advanced usage, please refer to our [documentation](https://medhaai.readthedocs.io).

## Contributing

Contributions are welcome! Please see our [Contributing Guide](CONTRIBUTING.md) for more details.

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
