ClinicalTrials.gov MCP Server

An MCP (Model Context Protocol) server that exposes the ClinicalTrials.gov v2 API as tools for LLM agents. Supports searching, retrieving, and analyzing clinical study data from the full registry of 570,000+ studies.

Tools

clinicaltrials_get_study

Retrieve a single study record by NCT ID.

Parameter	Type	Required	Description
nct_id	string	Yes	NCT identifier, e.g. NCT00000102
fields	StudyField[]	No	Specific fields to return; omit for all fields
format	json | csv	No	Response format (default: json)
markup_format	markdown | legacy	No	Text markup format (default: markdown)

clinicaltrials_search_studies

Search studies by condition, intervention, sponsor, location, and more. Returns paginated results.

Query parameters (at least one required):

Parameter	Description
query_cond	Condition or disease (e.g. diabetes)
query_term	General keyword search; also matches enum-typed fields (Phase, StudyType, InterventionType, etc.) when exact enum values are provided
query_intr	Intervention or treatment name; also matches InterventionType and ArmGroupType enum values
query_titles	Search within study titles only
query_id	NCT ID or other study identifier
query_spons	Sponsor or collaborator name
query_locn	Location search (text or AREA syntax: AREA[LocationState]MA, AREA[LocationCountry]US)
query_patient	Plain-language patient-friendly search; also matches StandardAge, InterventionalAssignment, DesignMasking, WhoMasked, ObservationalModel, PrimaryPurpose, and DesignTimePerspective enum values

Filters:

Parameter	Type	Description
filter_overall_status	OverallStatus[]	Recruitment status (e.g. [RECRUITING, NOT_YET_RECRUITING])
filter_geo	string	Geographic radius, e.g. distance(39.0,-77.0,50mi)
filter_ids	string[]	List of specific NCT IDs
post_filter_overall_status	OverallStatus[]	Status filter applied after aggregation counts
post_filter_geo	string	Geo filter applied after aggregation counts
agg_filter_phase	Phase[]	Filter by phase (e.g. [PHASE2, PHASE3])
agg_filter_study_type	StudyType[]	Filter by study type (e.g. [INTERVENTIONAL])
agg_filters	string	Raw aggregation filter string for advanced use; cannot be combined with agg_filter_phase or agg_filter_study_type

Pagination & output:

Parameter	Default	Description
sort	—	Sort field and direction, e.g. LastUpdatePostDate:desc
page_size	20	Results per page (1–1000)
page_token	—	Cursor from previous response's nextPageToken
count_total	false	Include total match count in response
fields	19 essential fields	Specific StudyField values to return. Defaults to essential fields (NCTId, BriefTitle, OverallStatus, Phase, Condition, InterventionName, sponsor, location, dates, enrollment). For comprehensive study details, use clinicaltrials_get_study instead.
format	json	json or csv
markup_format	markdown	markdown or legacy

clinicaltrials_search_datatable

Same search interface as clinicaltrials_search_studies but renders all matching results as an interactive, sortable, paginated table. Fetches all pages automatically. Each row shows NCT ID, title, status, phase, conditions, interventions, sponsor, start date, and enrollment count.

Accepts the same query, filter, and sort parameters as clinicaltrials_search_studies (excluding pagination and output format options).

clinicaltrials_get_field_values

Get the distribution of values across all studies for one or more fields. Useful for understanding what values exist and how common they are (e.g. study counts per phase, per country, per status).

Parameter	Type	Required	Description
fields	StudyField[]	Yes	Fields to get distributions for

Works best with enumerable fields: Phase, OverallStatus, StudyType, Sex, StdAge, LeadSponsorClass, InterventionType, LocationCountry, DesignAllocation, IsFDARegulatedDrug, HasResults, IPDSharing.

clinicaltrials_analyze_study_locations

Analyze the geographic distribution of locations across a set of studies. Pages through all matching results and classifies each study relative to a target country:

• only_in_target — all locations are in the target country
• mixed — locations in both the target country and at least one other
• not_in_target — no locations in the target country
• no_location_data — no location country data present

Also returns a frequency table of every country seen, sorted by count descending.

Accepts the same query and filter parameters as clinicaltrials_search_studies, plus:

Parameter	Default	Description
target_country	United States	Reference country for classification (case-insensitive)

Field Selection

Default Fields for Search

To optimize context window usage, clinicaltrials_search_studies returns 19 essential fields by default instead of all 394 fields:

Identification: NCTId, BriefTitle, OfficialTitle, Acronym
Status: OverallStatus, StartDate, PrimaryCompletionDate, CompletionDate, LastUpdatePostDate
Core Info: BriefSummary, Condition, Phase, StudyType, EnrollmentCount, InterventionName
Sponsor: LeadSponsorName, LeadSponsorClass
Location: LocationCountry, LocationFacility

This provides sufficient information for initial study screening and filtering while keeping responses compact.

Getting Full Study Details

For comprehensive information on specific studies including all 394 fields, use clinicaltrials_get_study with a single NCT ID. This tool is designed for detailed study review after initial screening.

Custom Field Selection

All tools that return study records accept a fields parameter for custom field lists. The StudyField enum contains all 394 valid field names sourced from the ClinicalTrials.gov metadata API, grouped by section:

• Identification: NCTId, BriefTitle, OfficialTitle, Acronym, OrgFullName, ...
• Status: OverallStatus, StartDate, CompletionDate, LastUpdatePostDate, ...
• Sponsor: LeadSponsorName, LeadSponsorClass, CollaboratorName, ...
• Design: Phase, StudyType, DesignAllocation, DesignMasking, EnrollmentCount, ...
• Arms / Interventions: InterventionType, InterventionName, ArmGroupLabel, ...
• Outcomes: PrimaryOutcomeMeasure, SecondaryOutcomeMeasure, ...
• Eligibility: EligibilityCriteria, Sex, MinimumAge, MaximumAge, StdAge, ...
• Locations: LocationFacility, LocationCity, LocationState, LocationCountry, ...
• Results: participant flow, baseline characteristics, outcome measures, adverse events
• MeSH / Browse: ConditionMeshTerm, InterventionMeshTerm, ...

Suggested Queries

Phase and study type filters

• Find recruiting Phase 2 and Phase 3 interventional studies for breast cancer.
• Search for recruiting observational studies on diabetes.
• Are there any expanded access or temporarily unavailable pembrolizumab studies?

Intervention type

• Find placebo-controlled drug trials for type 2 diabetes that are currently recruiting.
• Show me biological interventions for melanoma in phase 2 or 3.
• What device feasibility studies are enrolling by invitation?

Study design and patient eligibility

• Find prospective cohort studies on long COVID that accept adult and older adult participants.
• Show me double-blind parallel assignment prevention trials for cardiovascular disease.
• What treatment studies for childhood asthma use a crossover design and are currently recruiting?

Location analysis

• How many recruiting diabetes trials are US-only?
• Of all NIH-sponsored studies, which are exclusively in the United States?
• Analyze the country distribution of phase 3 oncology interventional trials.

Setup

Prerequisites

• Python 3.13+
• uv

Install

git clone https://github.com/your-org/nih-clinicaltrials-mcp-server
cd nih-clinicaltrials-mcp-server
uv sync
uv pip install -e .

Run locally (stdio)

uv run python src/clinicaltrials/app.py

Run as HTTP server

PORT=8000 uv run python src/clinicaltrials/app.py
# or
uv run uvicorn clinicaltrials.app:app --port 8000

Connect from Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "clinicaltrials": {
      "command": "uv",
      "args": [
        "run",
        "--directory", "/path/to/nih-clinicaltrials-mcp-server",
        "python", "src/clinicaltrials/app.py"
      ]
    }
  }
}

Project Structure

src/clinicaltrials/
├── app.py            # FastMCP server init and transport config
├── tools/            # Tool implementations (one file per tool)
├── models/           # Pydantic models and enums
│   ├── __init__.py                          # Re-exports all classes
│   ├── enums.py                             # All enum types (Phase, StudyType, etc.)
│   ├── fields.py                            # StudyField enum (392 ClinicalTrials.gov fields)
│   ├── search.py                            # Base search/query classes
│   ├── get_study.py                         # GetStudyInput model
│   ├── search_studies_input.py              # SearchStudiesInput model
│   ├── get_field_values_input.py            # GetFieldValuesInput model
│   ├── analyze_study_locations_input.py     # AnalyzeStudyLocationsInput model
│   └── search_datatable_input.py            # SearchDatatableInput model
├── utils.py          # Shared HTTP client (Chrome TLS impersonation) and error handling
├── prompts.py        # MCP prompts
└── routes.py         # Custom HTTP routes (/health)

The models/ directory organizes 1100+ lines of Pydantic models into focused modules for improved readability while maintaining full backward compatibility with from clinicaltrials.models import ... imports.

Implementation Notes

TLS fingerprinting: The ClinicalTrials.gov CDN blocks standard Python HTTP clients via JA3 TLS fingerprint detection. This server uses curl-cffi to impersonate Chrome's TLS handshake, which resolves the 403 errors that httpx and requests produce.

No authentication required: The ClinicalTrials.gov API is public and requires no API key.

API version: ClinicalTrials.gov v2 (https://clinicaltrials.gov/api/v2). Data is refreshed daily.