Metadata-Version: 2.4
Name: agent-studio
Version: 0.1.1
Summary: No-code AI agent creation platform with Streamlit UI and FastAPI backend
Project-URL: Documentation, https://github.com/username/agent-studio#readme
Project-URL: Source, https://github.com/username/agent-studio
Author: Harish Santhanalakshmi Ganesan
License: MIT
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.11
Requires-Dist: anthropic>=0.37.1
Requires-Dist: asyncio>=3.4.3
Requires-Dist: build>=1.2.2.post1
Requires-Dist: chardet>=5.2.0
Requires-Dist: click>=8.1.0
Requires-Dist: fastapi>=0.115.2
Requires-Dist: filelock>=3.17.0
Requires-Dist: hawkins-agent<=0.1.7
Requires-Dist: hawkinsdb>=1.0.1
Requires-Dist: langchain>=0.3.18
Requires-Dist: openai>=1.62.0
Requires-Dist: pip>=25.0.1
Requires-Dist: psutil>=5.9.8
Requires-Dist: pypdf2>=3.0.1
Requires-Dist: python-multipart>=0.0.20
Requires-Dist: requests>=2.32.3
Requires-Dist: streamlit-agraph>=0.0.45
Requires-Dist: streamlit-flow-component
Requires-Dist: streamlit>=1.42.0
Requires-Dist: trafilatura>=2.0.0
Requires-Dist: twilio>=9.4.5
Requires-Dist: twine>=6.1.0
Requires-Dist: uvicorn>=0.34.0
Description-Content-Type: text/markdown


### Agent Studio


```

## Installation

1. Install the package from PyPI:
```bash
pip install agent-studio
```

## CLI Usage

The package provides two main CLI commands:

1. Start the AgentStudio application:
```bash
agentstudio start
```
This command starts both the Streamlit interface and FastAPI backend servers.

2. Check the version:
```bash
agentstudio version
```

**Important Note:** If you're a first-time user, we strongly recommend checking the tutorial webpage in the application after installation. The tutorial will guide you through building agents and workflows effectively.

## API Endpoints

### Process Workflow
- **URL**: `/api/process`
- **Method**: `POST`
- **Body**:
```json
{
    "content": "Your input data here"
}
```

### Health Check
- **URL**: `/health`
- **Method**: `GET`

## Example Usage

```bash
# Check health
curl http://localhost:8000/health

# Process workflow
curl -X POST http://localhost:8000/api/process \
     -H "Content-Type: application/json" \
     -d '{"content": "Example input data"}'
```

## Extending the Workflow

To add new workflow steps, create a new `WorkflowStep` instance:

```python
new_step = WorkflowStep(
    name="step_name",
    description="Step description",
    requires=["dependent_step_name"]
)
workflow_manager.add_step(new_step)
```

## Error Handling

The API includes comprehensive error handling and logging:
- All errors are logged with timestamps and context
- HTTP exceptions include detailed error messages
- Step execution failures are properly propagated

## License

MIT License 

## Author

Made with love by Harish Sanathanalakshmi Ganesan