ClinicalTrials

Bridges AI assistants to ClinicalTrials.gov data with search, details, and CSV tooling via MCP.

python

13

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "jackkuo666-clinicaltrials-mcp-server": {
      "command": "python",
      "args": [
        "clinical_trials_server.py"
      ]
    }
  }
}

The ClinicalTrials MCP Server connects AI assistants to ClinicalTrials.gov data through the Model Context Protocol, enabling fast trial searches and rich metadata access. It saves trial data locally, supports CSV workflows, and provides ready-made tools to retrieve, filter, and summarize clinical trial information for research and analysis.

How to use

You interact with the ClinicalTrials MCP Server through an MCP client integrated in your AI workflow. Start the local server, then invoke tools to search trials, fetch full study details, save results to CSV, and retrieve statistics. Use clear prompts like asking for trials related to a keyword, or requesting complete details for a specific NCT ID. The server exposes a set of tools you can call from your AI assistant to perform these tasks without hand-editing data or making separate API calls.

Typical goals you can achieve with your MCP client include searching for clinical trials with a custom expression, generating CSV files of study data, retrieving full study details by NCT ID, and obtaining summary statistics about trials. You can also load and inspect previously saved CSV files or list all available CSVs to manage your data locally.

How to install

Prerequisites: you need Python 3.10 or newer installed on your system.

Install and run the MCP server locally using the Python entry point. Start the MCP server with the following command from the project directory where the server script resides.

python clinical_trials_server.py

Additional setup and notes

You can also run the server in development environments by cloning the project, creating a virtual environment, and installing dependencies. The typical workflow is to clone the repository, set up a virtual environment, and install requirements before starting the server.

Project files include a primary MCP server implementation and helper utilities for interacting with the ClinicalTrials.gov API. Dependencies include Python-based tooling and data libraries. Ensure you have network access to fetch trial data during searches.

Available tools

search_clinical_trials_and_save_studies_to_csv

Search clinical trials using a search expression, optionally saving the results to a CSV file with selectable fields.

get_full_study_details

Retrieve comprehensive details for a single trial identified by its NCT ID.

get_studies_by_keyword

Find studies that match a given keyword, with optional CSV export.

get_study_statistics

Return statistics about clinical trials, with optional condition-based filtering.

get_full_studies_and_save

Fetch full studies data for a search expression and save the results to a CSV file.

load_csv_data

Load and display data from a specified CSV file.

list_saved_csv_files

List all CSV files saved in the local workspace.