home / mcp / research assistant mcp server

Research Assistant MCP Server

Provides semantic search across your personal research library with MCP clients and a Streamlit dashboard for management.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "chakri01-research-assistant-mcp": {
      "command": "python",
      "args": [
        "/full/path/to/research-assistant-mcp/mcp_server/server.py"
      ]
    }
  }
}

You have a production-ready MCP (Model Context Protocol) server that enables semantic search across your personal research library. It indexes documents, supports conversational queries with MCP clients like Claude Desktop, and provides a Streamlit dashboard for management and metrics, delivering fast, accurate results to help you find, summarize, and connect information across your papers and notes.

How to use

Start the MCP server and connect your MCP client to perform semantic searches across your documents. You will query in natural language, get top results with sources and metadata, and refine results with follow-up questions. Use the Streamlit dashboard to monitor usage, upload new documents, and visualize metrics. Integrations with Claude Desktop or other MCP clients let you converse with your library to retrieve and summarize content from multiple sources.

How to install

Prerequisites: you need Python 3.11 or later, at least 2 GB of RAM, and Git installed on your system.

1) Clone the repository and set up the project directory.

2) Create and activate a virtual environment.

3) Install dependencies from the requirements file.

4) Install local embeddings to enable fast vector-based search.

5) Configure environment variables and keys as needed for your embeddings provider.

Additional setup steps

# Clone repository
git clone https://github.com/yourusername/research-assistant-mcp.git
cd research-assistant-mcp

# Create virtual environment
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Install local embeddings
pip install sentence-transformers

# Configure environment
cp .env.example .env
# Edit .env - add OPENAI_API_KEY if using OpenAI embeddings

Start and connect the MCP server

Start the MCP server with Python. This runs the server locally so you can connect your MCP clients, such as Claude Desktop, to perform semantic searches.

python mcp_server/server.py

Configure Claude Desktop

Add a MCP connection to Claude Desktop so you can query your library directly from that environment.

{
  "mcpServers": {
    "research-assistant": {
      "command": "python",
      "args": ["/full/path/to/research-assistant-mcp/mcp_server/server.py"],
      "env": {}
    }
  }
}

Launch Streamlit UI

Open the Streamlit dashboard to upload documents, search, and visualize metrics.

streamlit run ui/app.py
```
Opens at http://localhost:8501

Available tools

search_documents

Semantic search across your library returning top-k results with sources, scores, and metadata

get_document_summary

Provide a quick overview of a document including title, authors, keywords, and a preview

find_related_papers

Find documents related to a topic and return relevance scores