search_clinical_trials_and_save_studies_to_csvLABSCO SUMMARY
Registry searches that leave a file behind.
The search happens in conversation but the output is a CSV on disk, so a landscape scan can be picked up in a spreadsheet instead of retyped out of a chat log.
What it is
A Python server over the ClinicalTrials.gov registry. It takes a search expression — plain terms or the registry's advanced syntax — caps how many studies come back, and writes the matches straight to a local CSV.
What you get
- search_clinical_trials_and_save_studies_to_csv runs a search expression, caps the number of studies, and saves the results to a CSV file
- Trial metadata by NCT ID, so a specific trial can be pulled up by its identifier
- Saved CSVs can be listed and loaded again, so a search you ran last week does not have to be re-run
- Summary statistics across the trials you retrieved
Requirements
Python with uv. The README's install line names ClinicalTrials-mcp-server, which PyPI does not have, so run it from a clone: the client entry activates the clone's virtual environment and starts clinical_trials_server.py. No account and no key: ClinicalTrials.gov data is public.
Setup effort
Build from source — git clone https://github.com/JackKuo666/ClinicalTrials-MCP-Server.git
USE CASES
01
Scanning the trial landscape for a condition
One search expression returns the matching studies and writes them out in the same step.
02
Keeping a search you will need again
Results are written to a named CSV that can be listed and reloaded later, so the query does not have to be repeated.
03
Looking up one trial by its NCT ID
Metadata for a single registered trial, retrieved by identifier rather than by search.
04
Sizing a field before reading any protocols
Statistics across the studies you pulled answer how many, how recent and how spread out.
7 tools
Listed in project docs. Your client may show a different list.
get_full_study_detailsget_studies_by_keywordget_study_statisticsget_full_studies_and_saveload_csv_datalist_saved_csv_filesBEFORE YOU INSTALL
What to know first
Tools you get
7 tools named in the project's documentation.
Requires
Python 3.10+
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
# Clone and set up development environment
git clone https://github.com/JackKuo666/ClinicalTrials-MCP-Server.git
cd ClinicalTrials-MCP-Server
# Create and activate virtual environment
uv venv
source .venv/bin/activate
uv pip install -r requirements.txtMCP Clients
Cline
CLINE · JSON
{
"mcpServers": {
"ClinicalTrials": {
"command": "bash",
"args": [
"-c",
"source /home/<YOUR_USERNAME>/PATH/ClinicalTrials-MCP-Server/.venv/bin/activate && python /home/<YOUR_USERNAME>/PATH/ClinicalTrials-MCP-Server/clinical_trials_server.py"
],
"env": {},
"disabled": false,
"autoApprove": []
}
}
}
