Metadata-Version: 2.4
Name: augagent
Version: 1.0.0
Summary: An enterprise-grade autonomous ReAct agent framework with token streaming and hierarchical orchestration.
Home-page: https://github.com/augmencord/augagent
Author: Augmencord Research
Author-email: research@augmencord.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: pydantic>=2.0.0
Requires-Dist: httpx>=0.24.0
Requires-Dist: chromadb>=0.4.0
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# AugAgent

An enterprise-grade autonomous ReAct agent framework with token streaming, hierarchical orchestration, and vector memory.

## Features
- **Real-Time Token Streaming**: Built heavily on `httpx.AsyncClient` for blazing fast, token-by-token UI updates.
- **Hierarchical Orchestration**: Includes `DelegateWorkTool` out-of-the-box, allowing "Manager" agents to spin up specialized sub-agents.
- **RAG & Memory**: Natively integrates ChromaDB for sliding-window and long-term vector memory abstractions.
- **Human-in-the-Loop**: Supports `require_human_approval` to yield payloads to an external UI for approval before code execution.
- **Provider Agnostic**: Use OpenAI, Anthropic, or local open-source models via Ollama.

## Installation
```bash
pip install augagent
```

## Quick Start
```python
import asyncio
from augagent import Agent, LLMConfig

async def main():
    config = LLMConfig(model="qwen2.5-coder:7b")
    agent = Agent(
        name="AugHome-Root", 
        role="Senior AI Engineer", 
        goal="Handle IDE requests", 
        llm_config=config
    )
    
    result = await agent.execute("Write a python script to calculate fibonacci numbers.")
    print(result.output)

if __name__ == "__main__":
    asyncio.run(main())
```
