ClinicalTrials
MCP server from BACH-AI-Tools/ClinicalTrials-MCP-Server
claude mcp add --transport stdio bach-ai-tools-clinicaltrials-mcp-server npx -y bach-clinical-trials
How to use
The Clinical Trials MCP Server exposes the full ClinicalTrials.gov API via MCP tools, enabling programmatic access to trial data within an MCP-enabled environment. It provides 18 specialized tools for searching, retrieving, and analyzing trials, including core search tools (e.g., search_studies, get_study_details, search_by_condition, search_by_location, search_by_sponsor, search_by_intervention) and advanced/auxiliary capabilities (get_similar_studies, get_trial_statistics, get_recruiting_studies, search_by_date_range, get_studies_with_results, search_rare_diseases, get_pediatric_studies, search_international_studies). Use these tools from your MCP client to query trials by condition, location, sponsor, phase, eligibility criteria, and more, and to fetch detailed metadata, timelines, and statistics. The API delivers structured JSON data including NCT IDs, study metadata, locations with coordinates, eligibility criteria, contacts, and data quality validations. To use it, configure an MCP server entry (as shown in the quickstart) and call the desired tool with appropriate parameters (for example, a condition and phase for search_studies, or an NCT ID for get_study_details). The server also supports auto-approval of commonly used tools in a claude_desktop_config.json setup, streamlining workflows for frequent queries.
How to install
Prerequisites:\n- Node.js 16+ installed on your system (from nodejs.org)\n- No API key required for ClinicalTrials.gov public API\n\nInstallation steps:\n1) Ensure you have npm/yarn available and a workspace for MCP servers.\n2) Run the quick start npx command provided by the project to fetch and run the server without local installation:\n\n npx -y bach-clinical-trials\n\n3) Alternatively, if you prefer a local install, clone the repository or install the npm package directly:\n\n npm install bach-clinical-trials\n npm run build (if the project ships TS sources)\n node path/to/clinical-trials-server/build/index.js (or the appropriate JS entry)\n\n4) Add the MCP configuration to Claude or your MCP client, for example:\n\n {\n "mcpServers": {\n "clinical-trials": {\n "command": "npx",\n "args": ["-y", "bach-clinical-trials"]\n }\n }\n }\n\n5) Verify that the server starts and is reachable by issuing a test tool call (e.g., search_studies) through your MCP client.\n\nNote: The quick start approach uses npx to fetch the latest version from npm, ensuring you run the most up-to-date server without manual installation. If you fork or host locally, ensure build artifacts exist under clinical-trials-server/build/ and that the entry point is index.js.
Additional notes
Tips and caveats:\n- The server updates its data daily on weekdays, so expect near-real-time results with occasional delays for high-traffic queries.\n- Ensure your claude_desktop_config.json (or equivalent) uses the correct path to the server build when referencing node path/to/clinical-trials-server/build/index.js.\n- NCT IDs follow the NCT######## format; the API performs regex validation to prevent invalid IDs.\n- If you encounter API or network errors, check for timeout protections and rate-limiting handling described in the documentation.\n- The npm package bach-clinical-trials is the recommended delivery method; using npx defaults to pulling the latest compatible server, reducing maintenance overhead.\n- When using autoApprove in configurations, only enable tools you actually use to minimize unnecessary data exposure.\n- Review the API reference and response formats to parse and display fields like location coordinates, sponsor information, and eligibility criteria correctly in your application.
Related MCP Servers
zen
Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage
MCP -Deepseek_R1
A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)
mcp-fhir
A Model Context Protocol implementation for FHIR
mcp
Inkdrop Model Context Protocol Server
mcp-appium-gestures
This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..
dubco -npm
The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.