What you can do with Clinical Trials

Built for

Patient Trial Search, Biotech Research, Pharma Pipeline Analysis, Investors

Example workflows

Find recruiting trials nearby

Patient path: condition, recruiting status, geography, then detail.

Try this

Find recruiting type 2 diabetes trials within 50 miles of Boston, open the best matches for full eligibility criteria, and list facility locations and study contacts.

Count Phase 3 trials

Sizing questions use count with filters — not a broad search.

Try this

How many Phase 3 breast cancer trials are currently recruiting on ClinicalTrials.gov?

Map a drug pipeline

Intervention search for pharma and investor pipeline scans.

Try this

Search ClinicalTrials.gov for pembrolizumab trials, group them by phase and status, and highlight the largest active studies by enrollment.

Studies with posted results

Separates registration metadata from posted outcome data.

Try this

Find completed GLP-1 agonist trials with posted results on ClinicalTrials.gov, sort by enrollment, and summarize primary outcomes.

Context to know first

Do I need extra setup to search ClinicalTrials.gov?

No. MCPBundles searches ClinicalTrials.gov live on each request — connect the server and start searching from chat.

What can I search on ClinicalTrials.gov from chat?

Conditions, interventions, sponsors, phases, recruitment status, locations, eligibility, facility contacts, study design, and posted results when sponsors have uploaded them.

Can patients use this to find recruiting trials?

Yes — filter by condition, recruiting status, and location, then read eligibility and contacts. This is not medical advice; confirm fit with a clinician or study coordinator.

Related editorial

ClinicalTrials.gov Search with AI

How structured trial data supports research by condition, sponsor, status, and study details.

Read article

AI Skill
SKILL.md

Domain knowledge for Clinical Trials — workflow patterns, data models, and gotchas for your AI agent.

Clinical Trials

ClinicalTrials.gov registrations, updated daily. Filter by condition, intervention, sponsor, phase, status, and location.

Operational order

  1. Search — list matching trials (summaries). Default for examples, sponsors, or NCT IDs.
  2. Count — number only. Unfiltered → total_studies; with filters → total_count.
  3. Study detail — one NCT ID for full eligibility, locations, and contacts. Search first unless the user gave an NCT ID.

Common paths

  • Patient trial search: condition + RECRUITING + location or geo_filter → summaries include central contacts; open study detail for full eligibility.
  • Eligibility screening: dedicated eligibility search with sex, age limits, eligibility keywords (BRCA, HbA1c), location, geo.
  • Investigator / site discovery: search investigators by PI name, affiliation, facility, and therapeutic area.
  • Drug or sponsor scan: intervention or sponsor + phase/status; first_posted_min for newly registered trials.
  • Sizing: count with any filter combo — including geo, has_results, sponsor_class=INDUSTRY.
  • Outcome measures: compare what primary endpoints trials in an area use — for protocol or competitive questions.

Gotchas

  • Status and phase are UPPERCASE (RECRUITING, PHASE3). Common aliases like "Phase 3" are accepted.
  • Search returns lean rows with contacts and location preview — open study detail for full eligibility and all sites.
  • Pagination uses next_page_token. Geo: distance(lat,lon,dist). Dates: YYYY-MM-DD on start_date_min/max and first_posted_min/max.
  • Count returns one number — no sponsor breakdown; run separate filtered counts or aggregate from a small search page.
  • No registry-wide popularity metric. Sort reorders the current result set only.
  • Generic drug names beat brand names for intervention search. Patient next step: call the trial coordinator to confirm eligibility.

Where the data comes from

Tool calls query ClinicalTrials.gov API v2 in real time, flatten common study fields for search results, preserve NCT IDs for detail lookup, and expose count queries for market and pipeline sizing. Results are not medical advice.

ClinicalTrials.gov API v2
National Library of Medicine, National Institutes of Health

Structured study registration and results data for interventional, observational, and expanded-access clinical studies.

580,000+ registered studies
Source updated: Daily
We refresh: Real-time (queried from ClinicalTrials.gov API v2 per request)
JSON HTTP API
Global clinical studies registered with ClinicalTrials.gov
ClinicalTrials.gov records are submitted by sponsors and investigators. Study data can be incomplete, delayed, or updated after initial registration.
Trial search results are not medical advice. Patients should review eligibility and participation decisions with clinicians and study contacts.
Recruitment status, enrollment, completion dates, and posted results can change. Review the linked ClinicalTrials.gov study record before acting.

Source availability last verified 2026-05-02.

Tools in this Server (6)

Clinical Trials Count

Count matching ClinicalTrials.gov studies without returning rows — use for sizing and head-to-head sponsor comparisons. Supports condition, interventi...

Clinical Trials Get Study

Get full details for one clinical trial by NCT ID — eligibility criteria, all site locations with site contacts, central study contacts (phone/email),...

Clinical Trials Outcome Patterns

Compare clinical outcome measures registered on ClinicalTrials.gov — the primary and secondary endpoints trials plan to measure (e.g. HbA1c change, ov...

Clinical Trials Search

Search ClinicalTrials.gov and return lean trial summaries with NCT ID, status, phase, sponsor, enrollment, dates, central contacts, and location previ...

Clinical Trials Search Eligibility

Screen ClinicalTrials.gov trials by patient demographics and eligibility keywords — for coordinators matching patients to studies. Filter by condition...

Clinical Trials Search Investigators

Find principal investigators and research sites running trials in a therapeutic area. Search by PI name, institution affiliation, facility, condition,...

Frequently Asked Questions

What is the Clinical Trials MCP server?

Search and analyze 580,000+ clinical trials from ClinicalTrials.gov. Find studies by condition, drug, sponsor, phase, recruitment status, and location. Get full study details including eligibility criteria, outcomes, design, locations, and contacts. Count trials for market sizing. It provides 6 tools that AI agents can use through the Model Context Protocol (MCP).

How do I connect Clinical Trials to my AI agent?

Add the MCPBundles server URL to your MCP client configuration (Claude Desktop, Cursor, VS Code, etc.). The URL format is: https://mcp.mcpbundles.com/bundle/clinical-trials. Authentication is handled automatically.

How many tools does Clinical Trials provide?

Clinical Trials provides 6 tools that can be called by AI agents, along with a SKILL.md that gives your AI agent domain knowledge about when and how to use them.

What authentication does Clinical Trials require?

Clinical Trials uses open data APIs — no authentication required.

Do I need extra setup to search ClinicalTrials.gov?

No. MCPBundles searches ClinicalTrials.gov live on each request — connect the server and start searching from chat.

What can I search on ClinicalTrials.gov from chat?

Conditions, interventions, sponsors, phases, recruitment status, locations, eligibility, facility contacts, study design, and posted results when sponsors have uploaded them.

Can patients use this to find recruiting trials?

Yes — filter by condition, recruiting status, and location, then read eligibility and contacts. This is not medical advice; confirm fit with a clinician or study coordinator.

Can it rank the most popular clinical trials?

No registry-wide popularity or ranking metric exists. Sort options reorder the current search results only — for large trials, sort by enrollment within a filtered search.

What does has_results mean?

It means outcome data was posted to ClinicalTrials.gov. Many completed studies never post results — treat registration status and result availability separately.

Setup Instructions

Connect Clinical Trials to any MCP client in minutes

MCP URL
https://mcp.mcpbundles.com/bundle/clinical-trials

One-click install:

The link prefills the Add custom connector dialog — you still review the values and click Add, then Connect to complete OAuth.

Or add manually

  1. Open claude.ai → Settings → Connectors.
  2. Click the + button and choose Add custom connector.
  3. Set Name to Clinical Trials and paste the MCP URL into Remote MCP server URL.
  4. Click Add. Clinical Trials will appear under Not connected — select it and click Connect to complete OAuth.
Name: Clinical Trials
Remote MCP server URL: https://mcp.mcpbundles.com/bundle/clinical-trials
Authentication: OAuth

Custom connectors at claude.ai require a paid Claude plan (Pro, Max, Team, or Enterprise).

Other ways to use Clinical Trials

Same data, different audiences.

REST API

Call every tool from your own backend. OpenAPI spec + cURL examples.

Open API docs

Try Clinical Trials now

No API key or third-party login required. Chat with AI and run tools instantly.

Clinical Trials MCP Server & Skill — 6 Tools