Update README.md

README.md

@@ -13,15 +13,103 @@ short_description: Streamlit template space
 
 # AI-Powered Scientific Research Companion
 
-This repository implements an AI-driven research assistant using the Model Context Protocol (MCP). It enables end-to-end capabilities:
+Welcome to your AI-driven research assistant—hosted as a Hugging Face Space!
 
-- 🔍 **Discovery & Ingestion**: Search journals, preprint servers, and web via MCP servers (e.g. `pskill9/web-search`, `metatool-ai/metatool-app`).
-- 📓 **Reproducible Notebook**: Retrieve and run code snippets for core experiments via a Python-run MCP server.
-- 🔗 **Knowledge Graph**: Build and visualize relationships (methods, datasets, concepts) via a vector-store MCP (e.g. Chroma).
-- 🔧 **Provenance Tracking**: Store metadata and run history in a SQL database via MCP.
+---
+
+## 🚀 Live Demo
+
+👉 https://huggingface.co/spaces/your-username/MCP_Research
+
+---
+
+## Features
+
+- 🔍 **One-Click Discovery**  
+  Search academic articles, preprints, and the open web via MCP servers (`pskill9/web-search`, `metatool-ai/metatool-app`).
+
+- 📓 **Reproducible Notebook**  
+  Fetch and run code snippets for core experiments using a Python-run MCP sandbox.
+
+- 🔗 **Interactive Knowledge Graph**  
+  Build and explore concept‐and‐method graphs via a vector-store MCP (Chroma).
+
+- 🛠️ **Provenance Tracking**  
+  Store metadata and execution history in a SQL database via MCP.
+
+- 🎨 **Light & Dark Themes**  
+  Toggle UI themes for day/night readability.
+
+---
 
-## Getting Started
-1. Clone this repo:
+## 🔧 Setup
+
+1. **Fork or Clone**  
    ```bash
-   git clone https://huggingface.co/spaces/your-username/MCP_Research
-   cd MCP_Research
+   git clone https://huggingface.co/spaces/your-username/MCP_Research.git
+   cd MCP_Research
+Configure Endpoints
+Copy and edit config.example.yaml → config.yaml:
+
+yaml
+Copy code
+mcp_servers:
+  web_search:  http://localhost:3000
+  pubmed:      http://localhost:3001
+  chroma:      http://localhost:3002
+  python_run:  http://localhost:3003
+db_url:       sqlite:///embeddings.db
+Install Dependencies
+
+bash
+Copy code
+pip install -r requirements.txt
+Run Locally (optional)
+
+bash
+Copy code
+streamlit run app.py --server.port 8501 --server.address 0.0.0.0
+📦 Deployment on Hugging Face
+Create a New Space
+
+Choose Streamlit as the runtime.
+
+Push your repo to the Space.
+
+Environment Variables / Secrets
+In your Space’s Settings → Secrets, add any API keys or credentials required by your MCP endpoints.
+
+Automatic Build & Launch
+HF will install from requirements.txt and launch app.py automatically.
+
+📁 Project Layout
+arduino
+Copy code
+├── .gitignore
+├── README.md
+├── config.yaml
+├── requirements.txt
+├── Dockerfile
+├── app.py
+├── orchestrator/
+│   ├── client.py
+│   ├── dispatcher.py
+│   └── provenance.py
+├── components/
+│   ├── sidebar.py
+│   ├── paper_list.py
+│   ├── notebook_view.py
+│   └── graph_view.py
+├── scripts/
+│   ├── ingest.py
+│   └── reindex.py
+└── demo.ipynb
+🤝 Contribute
+⭐ Star this Space
+
+🐛 Report issues or suggest features
+
+🔀 Submit PRs — we welcome enhancements and new MCP integrations!
+
+📜 License
+MIT © 2025