Add Hugging Face space config

README.md

@@ -1,135 +1,34 @@
-<<<<<<< HEAD
 ---
-title: Echo Pilot
-=======
-<<<<<<< HEAD
----
-title: EchoPilot
->>>>>>> 3e784b6 (init commit)
-emoji: 🚀
-colorFrom: red
+title: Echo (Private EchoPilot Agent)
+emoji: 🫀
+colorFrom: blue
 colorTo: red
-sdk: docker
-app_port: 8501
-tags:
-- streamlit
+sdk: streamlit
+sdk_version: "1.39.0"
+app_file: app.py
 pinned: false
-<<<<<<< HEAD
-short_description: Echo-Pilot space
-=======
-short_description: The Echocardiography Agent Framework
->>>>>>> 3e784b6 (init commit)
-license: mit
+private: true
 ---
 
-# Welcome to Streamlit!
-
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
-
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-<<<<<<< HEAD
-=======
-=======
-# EchoPilot Agent
-
-An echocardiography ReAct agent focusing on EchoPilot’s specialised tools.
+# EchoPilot – Private Echocardiography Agent
 
-## Project Structure
+This Hugging Face Space runs the EchoPilot Streamlit interface for cardiac ultrasound analysis (segmentation, measurements, disease prediction, view classification, reporting, and video generation).
 
+## Usage
+1. Open the Space (private link only).
+2. Provide the echo video path and clinical question.
+3. The ReAct agent orchestrates the available tools and responds with a structured summary.
 
-```
-echo-agent/
-├── agents/                    # Agent implementations
-│   ├── intelligent_agent.py   # ReAct loop with tool calling
-│   └── react_agent.py         # LangGraph workflow wrapper
-├── assets/                    # Static assets and data files
-├── configs/                   # Configuration files
-├── docs/                      # Documentation
-├── model_weights/             # Pre-trained weights
-├── models/                    # Model wrappers and factory
-├── outputs/                   # Generated outputs
-├── scripts/                   # Utility scripts
-├── temp/                      # Temporary files
-├── tests/                     # Test suite
-├── tool_repos/                # External tool repositories
-│   ├── MedSAM2-main
-│   └── EchoFlow
-├── tools/                     # Tool implementations
-├── utils/                     # Utility helpers
-├── config.py                  # Main configuration
-├── main.py                    # CLI entry point
-├── README.md                  # Project documentation
-├── requirements.txt           # Python dependencies
-├── setup.py                   # Package setup
-└── streamlit_app.py           # Streamlit web interface
-```
-
-## 🚀 Quick Start
+## Secrets & Config
+Configure the following in **Settings → Variables and secrets**:
+- `OPENAI_API_KEY`
+- `ECHO_INITIAL_MASK_PATH` (default mask prompt, optional)
+- Any other environment variables from `config.py`
 
-### 1. Launch the Streamlit interface
+## Local Development
 ```bash
+pip install -r requirements.txt
 streamlit run streamlit_app.py
 ```
 
-### 2. Run the CLI agent
-```bash
-python -m echo-agent.main --video path/to/video.mp4 --query "Assess LV systolic function."
-```
-
-### 3. Run Comprehensive Tests
-```bash
-python tests/test_all_tools_comprehensive.py
-```
-
-## 🔧 Key Components
-
-### Query Analyzer System
-- **API Server**: REST API for intelligent tool selection
-- **Client Library**: Python client with fallback support
-- **Hybrid Analyzer**: Automatic API/local switching
-
-### AI Models
-- **PanEcho**: Cardiac disease prediction (40 tasks)
-- **MedSAM2**: Medical image segmentation
-- **EchoFlow**: Video generation and analysis
-- **EchoPrime**: Advanced echocardiogram analysis
-
-### Web Interfaces
-- **Streamlit**: Primary web interface with progress tracking
-
-
-### Tools
-- **Echo Segmentation**: Cardiac structure segmentation
-- **Disease Prediction**: Comprehensive cardiac analysis
-- **View Classification**: Echocardiogram view identification
-- **Measurement Prediction**: Cardiac measurements
-- **Report Generation**: Clinical report creation
-- **Video Generation**: Synthetic echocardiogram creation
-
-## 🛠️ Development
-
-### Adding New Tools
-1. Implement tool in `tools/` directory
-2. Ensure the tool name and description appear in `agents/intelligent_agent.py` prompt if needed
-3. Update web interfaces or CLI helpers if surfacing new outputs
-4. Add tests
-
-### Adding New Models
-1. Implement model in `models/` directory
-2. Add to `model_factory.py`
-3. Update tool implementations
-4. Add configuration files
-
-
-## 📝 Configuration
-
-All configuration is managed through:
-- `config.py`: Main configuration
-- `configs/`: Model-specific configurations
-- Environment variables for deployment
-
-
-
->>>>>>> 94e3313 (Initial EchoPilot Space)
->>>>>>> 3e784b6 (init commit)
+Please keep PHI and large datasets out of this Space; only the runnable code and lightweight assets are included. Weight files download at runtime or reside in the bundled `tool_repos/` directories.