Pharma, FDA & ClinicalTrials.gov Intelligence MCP

io.github.willowswt由 GitHub 帳號 willowswt 發布v1.0.0更新於 Nov 19, 2026

ClinicalTrials.gov v2 studies, OpenFDA FAERS adverse events & recalls, and PubMed evidence.

發布方自己的服務:[PREPROD] Clinical.ly Research Suite

已驗證Streamable HTTP可網頁執行金融知識與記憶資料與分析

概覽

AI 產生的概覽

遠端 MCP 伺服器,讓助理以精簡、可引用的形式查詢 ClinicalTrials.gov 試驗、OpenFDA 安全與召回資料,以及 PubMed 文獻證據。

可以做什麼
  • 搜尋與讀取臨床試驗:依適應症、藥物、贊助者、期別或狀態尋找研究,並取得含納入排除條件、給藥組與終點的完整 NCT 計畫
  • 分析競爭產品線:將某適應症的試驗彙整為期別與狀態分布、主要贊助者、分子及即將到來的讀出
  • 查詢 FDA 安全資料:取得 Drugs@FDA 核准資訊、FAERS 不良反應計數與嚴重結果,以及藥品強制召回
  • 搜尋 PubMed 文獻:回傳同儕審查證據的 PMID、期刊、DOI、作者與日期
適用情境
適合生技醫藥盡職調查、競爭格局備忘錄、藥物警戒與召回監測,或任何需要以臨床試驗、法規與文獻原始來源為依據、而非依賴模型既有知識的問題。
執行需求
遠端 Streamable HTTP 端點,無需本機安裝。需要 Apify 帳號,並以 Authorization 標頭中的 bearer 形式提供 Apify API 權杖。不需 FDA、NIH 或 PubMed 金鑰。工具呼叫依 Apify 按事件計費,含少量每月免費額度。
安裝前請注意
每次工具呼叫都會計入 Apify 帳號費用,反覆或由代理自動發起的查詢可能消耗額度;免費方案每次呼叫僅回傳少量試驗紀錄。Authorization 標頭帶有你的 Apify API 權杖,請勿放入共用設定。輸出為公開登錄紀錄,僅供研究與盡職調查,不構成醫療建議。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Pharma, FDA & ClinicalTrials.gov Intelligence MCP,將其新增到工作區。
  2. 連接 Authorization token 憑證。它們會被私密儲存,絕不會在公開頁面上展示。
  3. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.willowswt%2Fpharma-fda-clinical-trials-mcp/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "pharma-fda-clinicaltrials-gov-intelligen": {
      "type": "http",
      "url": "https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

[Glama MCP Server] [Apify Store Actor] [MCP Protocol] [License: MIT]

⚡ Instant Remote Streamable-HTTP Endpoint (Zero Local Install): https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp
🔑 Get Your Free API Token ($5.00/mo Free Credit Included): Apify Console -> Settings -> Integrations
🌐 Hosted Apify Store Page & Interactive Runner: https://apify.com/church_of_loujess/pharma-fda-clinical-trials-mcp

What can Pharma, FDA & Clinical Trials MCP Server do?

Pharma, FDA & Clinical Trials MCP Server is a live, zero-API-key Streamable-HTTP Model Context Protocol (MCP) server running in 128 MB Standby Mode (https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp). Connect Claude Desktop, Cursor, ChatGPT Developer Mode, Windsurf, Codex, or custom AI agents to 5 authoritative regulatory & biomedical tools spanning ClinicalTrials.gov API v2, US OpenFDA (Drugs@FDA, FAERS adverse events, enforcement recalls), and NCBI PubMed — or run any tool directly from Apify Console:

📦 What you can extract⚙️ Features & integrations

🧪 Live clinical trials (ClinicalTrials.gov v2) by condition, drug, sponsor, Phase 1–4, or recruitment status
📋 Full NCT protocol details inclusion/exclusion criteria, dosing arms, primary/secondary endpoints & site counts
📊 Competitive pipeline maps sponsor share, phase breakdown, total enrollment & nearest primary completion dates
⚠️ OpenFDA FAERS & recalls Drugs@FDA approvals, top MedDRA adverse reactions, serious outcomes & enforcement recalls
📚 NCBI PubMed peer-reviewed evidence PMIDs, journals, DOIs, publication types, lead/senior authors & dates
💸 $0.00 free MCP discovery initialize, tools/list, and ping are completely free; pay only per tools/call

⚡ Native Streamable HTTP (/mcp) supports MCP protocols 2025-06-18, 2025-03-26 & 2024-11-05
🔓 Zero external API keys no FDA, NIH, or PubMed keys needed — connect with just your Apify token
🪶 Ultra-fast 128 MB Standby sub-second warm response with zero headless browser or proxy overhead
🔀 Dual-Mode execution works both as a live /mcp server AND as a standard Console/API Batch Actor
⬇️ Structured JSON + Dataset returns clean LLM-sized payloads (aggregating 2.3 MB raw FAERS blobs automatically)
🤖 Plug-and-play AI clients ready for Claude, Cursor, Windsurf, OpenAI Responses API, LangGraph & CrewAI

Biopharma BD&L teams, healthcare hedge funds, regulatory affairs specialists, medical writers, and clinical AI builders normally juggle three incompatible government APIs — including OpenFDA raw event payloads that exceed 2 MB per record and blow up LLM context windows. This MCP server unifies all three into compact, citation-ready tool responses.

What data can I extract with Pharma, FDA & Clinical Trials MCP Server?

Every MCP tools/call (and every Console Batch run) pulls live records directly from official US federal biomedical registries — zero cached stale snapshots:

  • ClinicalTrials.gov API v2 search & full NCT protocols: from indication scan to inclusion/exclusion criteria. search_clinical_trials and get_trial_details return nctId, title, status, phases, studyType, enrollment, leadSponsor, sponsorClass (INDUSTRY, NIH, OTHER), startDate, primaryCompletionDate, conditions, interventions, primaryOutcomes, eligibility criteria, and canonical https://clinicaltrials.gov/study/<NCT> URLs.

  • Indication-level competitive pipeline analytics: instant BD&L landscape tables. analyze_competitive_pipeline computes totalMatchingStudiesOnRegistry, phaseDistribution, statusDistribution, topSponsors, topInterventions, totalSampledEnrollment, and upcomingPrimaryReadouts.

  • Context-safe OpenFDA safety, approvals & recalls: server-side MedDRA reaction aggregation. Instead of dumping raw megabyte FAERS reports, detect_fda_adverse_events_and_recalls queries OpenFDA's server-side count=patient.reaction.reactionmeddrapt.exact endpoint plus drugsfda and drug/enforcement to deliver exact reaction frequencies, serious vs. non-serious counts, NDA/BLA numbers, and Class I/II/III recall notices in a crisp 4 KB JSON payload.

  • NCBI PubMed E-Utilities literature citations: peer-reviewed grounding with PMIDs & DOIs. search_pubmed_clinical_evidence resolves pmid, title, journal, pubDate, authors, doi, pubTypes, and https://pubmed.ncbi.nlm.nih.gov/<pmid>/.

🛠️ Live MCP Tools Included on This Server

MCP Tool NameArguments (params.arguments)What It Returns
search_clinical_trialscondition, intervention, sponsor, phase (PHASE1..PHASE4), status, maxResults (1–50)Search ClinicalTrials.gov v2 and return NCT IDs, titles, phases, enrollment, sponsors, completion dates, interventions & primary outcomes.
get_trial_detailsnctId (required, e.g. "NCT04280705")Fetch full study protocol, brief/detailed summary, eligibility criteria, study design, dosing arms, primary/secondary endpoints & site summary.
analyze_competitive_pipelinecondition, sponsor, maxStudiesSampled (10–50, default 35)Aggregate up to 50 live trials in an indication into phase counts, status breakdown, top biopharma sponsors, top molecules & upcoming readouts.
detect_fda_adverse_events_and_recallsdrugName (required, brand or generic e.g. "ozempic", "semaglutide")Query OpenFDA for Drugs@FDA NDA/BLA approvals, FAERS top MedDRA adverse reaction counts + serious outcome split, and FDA drug recalls.
search_pubmed_clinical_evidencequery (required), maxResults (1–25, default 10)Search NCBI PubMed biomedical literature and return PMIDs, article titles, journals, DOIs, authors, dates & direct PubMed URLs.

Need more specialized workflows? Combine this Actor with these dedicated tools from our suite:

How much will Pharma, FDA & Clinical Trials MCP Server cost?

Pharma, FDA & Clinical Trials MCP Server uses transparent Pay-Per-Event (PPE) pricing where one event (mcp-tool-call) equals one successful MCP Tool Call. Connecting your AI client (initialize), inspecting tool schemas (tools/list), and health/keepalive checks (ping) are $0.00 FREE — there is $0.00 Actor start fee on MCP calls.

Paid Apify plans automatically unlock volume discounts up to 52% off ($0.035 on Starter/Bronze, $0.028 on Scale/Silver, and $0.024 on Business/Gold, Platinum & Diamond) with zero residential proxy surcharges:

Apify Plan / TierPrice per MCP tools/call executionPrice per 1,000 MCP tool callsBatch Depth & UnmaskingApprox. Monthly Capacity
Free (FREE Trial Preview)$0.05$50.003 records/call trial (first 2 unmasked, 3rd preview-masked)~100 trial calls
Starter (BRONZE — $29/mo)$0.035$35.00Unlimited depth & 100% unmasked~828 MCP tool calls
Scale (SILVER — $199/mo)$0.028$28.00Unlimited depth & 100% unmasked~7,107 MCP tool calls
Business (GOLD — $999/mo)$0.024$24.00Unlimited depth & 100% unmasked~41,625 MCP tool calls
Enterprise (PLATINUM / DIAMOND)$0.024 / $0.024$24.00Unlimited depth & 100% unmaskedCustom high-volume SLA

Free Trial vs. Paid Plan Unlock: On the Apify Free tier, each mcp-tool-call returns up to 3 trial records (first 2 records 100% unmasked, 3rd record preview-masked). Upgrading to Apify Starter ($29/mo) or higher unlocks full multi-record batch depth, 100% unmasked records, and up to 52% lower per-call rates. Check the pricing tab for full details.

How do I use Pharma, FDA & Clinical Trials MCP Server?

Pharma, FDA & Clinical Trials MCP Server is engineered for zero-friction setup whether you are connecting an AI coding/chat client over Streamable HTTP (/mcp) or running batch extractions in Apify Console:

Option A: Connect to Claude Desktop, Cursor, Windsurf, or ChatGPT in 60 seconds

  1. Create a free Apify account and copy your Personal API Token from Console -> Settings -> Integrations.
  2. Add the live Streamable-HTTP endpoint https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp to your AI client's mcpServers config:
json
{  "mcpServers": {    "pharma-fda-clinical-trials-mcp": {      "url": "https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp",      "headers": {        "Authorization": "Bearer <YOUR_APIFY_API_TOKEN>"      }    }  }}
  1. Restart or reload your MCP client — it will automatically run initialize and tools/list ($0.00 free) and expose all live tools to your AI assistant.
  2. Ask your question in plain English; your AI agent will select the right tool, pass structured arguments, and cite live primary sources.

Option B: Run directly in Apify Console (No-Code Batch Mode)

  1. Open Pharma, FDA & Clinical Trials MCP Server in Apify Console.
  2. Choose which tool to execute under MCP Tool to Execute (Batch Mode) (tool).
  3. Fill in the matching input parameters (condition, intervention, sponsor, phase) in the Console input form.
  4. Click Save & Start and wait a few seconds for live extraction to finish.
  5. Preview your results in the Storage -> Dataset tab and export in JSON, CSV, Excel, or XML, or fetch via the Apify REST API.

⬇️ Input

Use this Actor in two ways: (1) Live Streamable-HTTP MCP Mode (POST https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp) from any AI client, OR (2) Console / REST Batch Mode by selecting tool and filling the matching parameters.

Mode 1: Live Streamable-HTTP MCP Client Connection (/mcp)

Paste into claude_desktop_config.json, .cursor/mcp.json, Windsurf, or OpenAI Agents SDK — your AI assistant automatically discovers all 5 tools ($0.00 handshake) and invokes them on demand.

Input / ConfigurationWhat it returns
search_clinical_trials ({"condition": "NASH", "phase": "PHASE3"})Active Phase 3 MASH/NASH trials, sponsors & primary readout dates
detect_fda_adverse_events_and_recalls ({"drugName": "ozempic"})Drugs@FDA approval + top FAERS MedDRA reactions + FDA recall history
search_pubmed_clinical_evidence ({"query": "tirzepatide heart failure"})Peer-reviewed PubMed citations with PMIDs & DOIs

Input JSON example:

json
{    "jsonrpc": "2.0",    "id": 1,    "method": "tools/call",    "params": {        "name": "detect_fda_adverse_events_and_recalls",        "arguments": {            "drugName": "semaglutide"        }    }}

Mode 2: Apify Console & REST API Batch Execution

Run any of the 5 tools directly as a standard Apify Actor run and export results to JSON, CSV, or Excel.

Input / ConfigurationWhat it returns
tool: "search_clinical_trials" + condition: "obesity" + phase: "PHASE3"Writes structured clinical_trial items into your default Apify Dataset
tool: "get_trial_details" + nctId: "NCT04280705"Full protocol record in Dataset Storage

Input JSON example:

json
{    "tool": "search_clinical_trials",    "condition": "obesity",    "intervention": "semaglutide",    "phase": "PHASE3",    "status": "RECRUITING",    "maxResults": 10}

Complete Input Parameter Reference

ParameterTypeDefaultRequiredDescription
toolstring (search_clinical_trials, get_trial_details, analyze_competitive_pipeline, detect_fda_adverse_events_and_recalls, search_pubmed_clinical_evidence)"search_clinical_trials"NoSelect which Pharma/FDA/PubMed MCP tool to execute when running directly from Apify Console or API.
conditionstring"obesity"NoMedical condition, indication, or therapeutic area (e.g. 'obesity', 'non-small cell lung cancer', 'alzheimer').
interventionstring—NoOptional drug, biologic, molecule, or device name (e.g. 'semaglutide', 'pembrolizumab').
sponsorstring—NoOptional lead sponsor name (e.g. 'Eli Lilly', 'Novo Nordisk', 'Pfizer').
phasestring—NoOptional ClinicalTrials.gov phase filter: PHASE1, PHASE2, PHASE3, PHASE4.
statusstring—NoOptional status filter: RECRUITING, ACTIVE_NOT_RECRUITING, COMPLETED, NOT_YET_RECRUITING.
nctIdstring—NoExact NCT identifier such as NCT04280705.
drugNamestring—NoBrand or generic drug name for FDA approval, FAERS adverse event, and recall intelligence (e.g. 'ozempic', 'semaglutide').
querystring—NoSearch query for peer-reviewed clinical studies on NCBI PubMed.
maxResultsinteger10NoMaximum records to return (1-50).

⬆️ Output and setup

All extracted records stream incrementally into your run's Dataset under the Storage tab in Apify Console. Preview results as a clean interactive table, download in JSON, JSONL, CSV, Excel (.xlsx), XML, or RSS, or consume items directly via the Apify REST API, Python/Node.js SDKs, or MCP.

📦 Verified dataset output sample #1

json
{    "tool": "search_clinical_trials",    "recordType": "clinical_trial",    "identifier": "NCT00006307",    "title": "Epidemiology of Body Mass Index Rebound",    "organization": "National Heart, Lung, and Blood Institute (NHLBI)",    "statusOrPhase": "N/A | COMPLETED",    "date": "2006-07",    "url": "https://clinicaltrials.gov/study/NCT00006307",    "details": {        "nctId": "NCT00006307",        "title": "Epidemiology of Body Mass Index Rebound",        "status": "COMPLETED",        "phases": [],        "studyType": "OBSERVATIONAL",        "enrollment": null,        "leadSponsor": "National Heart, Lung, and Blood Institute (NHLBI)",        "sponsorClass": "NIH",        "startDate": "2000-08",        "primaryCompletionDate": "2006-07",        "conditions": [            "Cardiovascular Diseases",            "Heart Diseases"        ],        "interventions": [],        "primaryOutcomes": [],        "hasPostedResults": false,        "url": "https://clinicaltrials.gov/study/NCT00006307"    }}

📦 Verified dataset output sample #2

json
{    "tool": "search_clinical_trials",    "recordType": "clinical_trial",    "identifier": "NCT01616732",    "title": "Testosterone and Weight Loss",    "organization": "Austin Health",    "statusOrPhase": "PHASE2 | COMPLETED",    "date": "2015-11",    "url": "https://clinicaltrials.gov/study/NCT01616732",    "details": {        "nctId": "NCT01616732",        "title": "Testosterone and Weight Loss",        "status": "COMPLETED",        "phases": [            "PHASE2"        ],        "studyType": "INTERVENTIONAL",        "enrollment": 100,        "leadSponsor": "Austin Health",        "sponsorClass": "OTHER_GOV",        "startDate": "2013-04",        "primaryCompletionDate": "2015-11",        "conditions": [            "Obesity in Men"        ],        "interventions": [            {                "type": "DRUG",                "name": "Testosterone",                "description": "TU im"            },            {                "type": "DRUG",                "name": "placebo",                "description": "placebo im"            }        ],        "primaryOutcomes": [            {                "measure": "weight",                "timeFrame": "12 months"            }        ],        "hasPostedResults": false,        "url": "https://clinicaltrials.gov/study/NCT01616732"    }}

Output Field Dictionary

Output FieldTypeDescription
tool / recordTypestringExecuted MCP tool name and normalized entity type (clinical_trial, fda_safety_report, pubmed_article, pipeline_summary).
identifierstringPrimary registry ID (NCT04280705, PMID:38912345, or drug name).
title / organizationstringStudy/article title and lead biopharma sponsor, manufacturer, or journal.
statusOrPhase / datestringPhase/recruitment summary (e.g. PHASE3 | RECRUITING) and completion/publication date.
urlstringDirect verification link on clinicaltrials.gov, open.fda.gov, or pubmed.ncbi.nlm.nih.gov.
detailsobjectComplete structured payload (interventions, endpoints, eligibility, FAERS reaction counts, NDA numbers, DOIs).

Notes, tips, and exceptions

  • $0.00 MCP handshake & discovery: initialize, notifications/initialized, tools/list, and ping are never billed — the mcp-tool-call event fires strictly once per successful tools/call execution.
  • Why 128 MB Standby matters: Running in 128 MB RAM keeps Standby compute overhead minimal so almost 100% of your Apify plan credit goes directly toward tool calls.
  • Valid phase & status enums: On ClinicalTrials.gov v2, pass uppercase EARLY_PHASE1, PHASE1, PHASE2, PHASE3, PHASE4, and statuses RECRUITING, ACTIVE_NOT_RECRUITING, NOT_YET_RECRUITING, COMPLETED.
  • Not medical advice: All outputs reflect public regulatory and scientific registry records (ClinicalTrials.gov, OpenFDA, NCBI PubMed) for research, regulatory intelligence, and due diligence — never clinical patient advice.

What can you do with data from Pharma, FDA & Clinical Trials MCP Server?

Pharma, FDA & Clinical Trials MCP Server powers high-ROI automation, lead generation, and AI agent pipelines. Here are four proven step-by-step recipes:

🧬 Biotech hedge-fund & BD&L indication landscape memo in Claude

Map every Phase 2/3 competitor, enrollment target, and upcoming catalyst date in 30 seconds.

Recipe:

  1. Connect /mcp in Claude Desktop or Cursor. Add the Standby URL + Authorization: Bearer <APIFY_TOKEN>.

  2. Prompt a multi-tool diligence workflow. Ask Claude: 'Map the Phase 2 and Phase 3 competitive pipeline in idiopathic pulmonary fibrosis, inspect protocols for the top 3 recruiting trials, and summarize recent PubMed clinical readouts.'

  3. Chain with SEC cash-runway checks. Add SEC EDGAR, UK Companies House & LEI MCP in the same Claude config to verify each biotech's cash balance and 10-Q burn rate!

Replaces hours of manual ClinicalTrials.gov + EDGAR tab-switching.

⚠️ Automated pharmacovigilance & FDA recall surveillance

Monitor post-market FAERS adverse event profiles and enforcement actions across a portfolio of approved drugs.

Recipe:

  1. Schedule weekly Batch or n8n runs. Call detect_fda_adverse_events_and_recalls for each drug in your formulary.

  2. Track serious outcome ratios & new recalls. Alert regulatory affairs teams on Slack whenever a new FDA enforcement recall appears.

Zero OpenFDA pagination or megabyte-JSON crashes.

📝 Grounded medical & scientific writing assistant

Prevent LLM citation hallucinations when drafting clinical backgrounders or grant sections.

Recipe:

  1. Invoke search_pubmed_clinical_evidence + get_trial_details. Force your agent to cite real PMID, DOI, and NCT identifiers verified live.

Every claim links directly to pubmed.ncbi.nlm.nih.gov or clinicaltrials.gov.

🤝 CRO & clinical-vendor B2B sales prospecting

Biotechs with NOT_YET_RECRUITING Phase 1/2/3 trials actively procure CROs, patient-recruitment agencies, eCOA, and cold-chain logistics.

Recipe:

  1. Query status: "NOT_YET_RECRUITING". Filter sponsorClass == "INDUSTRY".

  2. Enrich biotech sponsor contacts. Pass leadSponsor domains into Website Email, Phone & Tech Stack Finder.

High-precision trigger-based outbound for life-sciences B2B teams.

Want to get other B2B, Social, Financial, or MCP Agent data?

Chain this tool with our other 128 MB / 256 MB zero-proxy Apify Scrapers and Live Streamable-HTTP MCP Servers:

🔌 Using Pharma, FDA & Clinical Trials MCP Server with the Apify API, SDKs & MCP

Automate Pharma, FDA & Clinical Trials MCP Server programmatically from Python, Node.js/TypeScript, cURL, workflow builders (n8n, Make, Zapier, Clay, Pipedream), or Model Context Protocol (MCP) AI clients.

Python (apify-client)

python
from apify_client import ApifyClient
client = ApifyClient("<YOUR_APIFY_API_TOKEN>")run = client.actor("church_of_loujess/pharma-fda-clinical-trials-mcp").call(    run_input={    "tool": "search_clinical_trials",    "condition": "obesity",    "intervention": "semaglutide",    "phase": "PHASE3",    "status": "RECRUITING",    "maxResults": 10})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():    print(item)

Node.js / TypeScript (apify-client)

javascript
import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: '<YOUR_APIFY_API_TOKEN>' });const run = await client.actor('church_of_loujess/pharma-fda-clinical-trials-mcp').call({    "tool": "search_clinical_trials",    "condition": "obesity",    "intervention": "semaglutide",    "phase": "PHASE3",    "status": "RECRUITING",    "maxResults": 10});
const { items } = await client.dataset(run.defaultDatasetId).listItems();console.log(`Extracted ${items.length} items:`, items);

One-Line Synchronous REST API (cURL — ideal for n8n, Make & Clay HTTP Columns)

bash
curl -X POST "https://api.apify.com/v2/acts/church_of_loujess~pharma-fda-clinical-trials-mcp/run-sync-get-dataset-items?token=<YOUR_APIFY_API_TOKEN>" \  -H "Content-Type: application/json" \  -d '{"tool": "search_clinical_trials", "condition": "obesity", "intervention": "semaglutide", "phase": "PHASE3", "status": "RECRUITING", "maxResults": 10}'

Claude Desktop, Cursor, Windsurf & OpenAI MCP Configuration

json
{  "mcpServers": {    "pharma-fda-clinical-trials-mcp": {      "url": "https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp",      "headers": {        "Authorization": "Bearer <YOUR_APIFY_API_TOKEN>"      }    }  }}
Claude Code CLI one-liner:
bash
claude mcp add --transport http pharma-fda-clinical-trials-mcp "https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp" --header "Authorization: Bearer <YOUR_APIFY_API_TOKEN>"

❓ Frequently Asked Questions

Am I charged when Claude or Cursor connects and lists the tools (tools/list)?

No! initialize, notifications/initialized, tools/list, and ping are $0.00 FREE. You are only charged once per successful tools/call invocation ($0.05 on Free down to $0.024 on Gold–Diamond).

Do I need an OpenFDA API key or NCBI Entrez key?

No — zero third-party keys are required. Only your standard APIFY_API_TOKEN is used to authenticate against the Standby /mcp endpoint.

How do I connect this MCP server to Claude Desktop, Cursor, or Windsurf?

Add "url": "https://church-of-loujess--pharma-fda-clinical-trials-mcp.apify.actor/mcp" with header "Authorization": "Bearer <YOUR_APIFY_API_TOKEN>" in your mcpServers config (exact copy-paste JSON is in the integration section above).

Can I also run it without MCP as a normal Apify Actor or REST API?

Yes! In Apify Console or via apify-client, simply select tool (for example "search_clinical_trials") and pass condition, drugName, nctId, or query. Results save directly to your Apify Dataset.

Why doesn't detect_fda_adverse_events_and_recalls overflow Claude's context window?

Raw OpenFDA /drug/event.json records often exceed 2.3 MB each because patient medication lists include thousands of concomitant entries. Our server executes server-side MedDRA reaction count aggregations + targeted NDA/recall queries so responses stay compact (~3–6 KB) and accurate.

Are FAERS adverse event counts proof of causation?

No — per FDA guidance, FAERS reflects spontaneous post-market reports and does not by itself establish causal incidence rates.

Is it legal to extract public data with Pharma, FDA & Clinical Trials MCP Server?

Pharma, FDA & Clinical Trials MCP Server only accesses publicly available web pages, feeds, and registries that require no private account login or authentication bypass. Keep in mind that extracted contact fields (such as public business emails, names, or usernames) can still qualify as personal data under frameworks like the GDPR or CCPA depending on jurisdiction. Always ensure you have a lawful basis (such as legitimate B2B interest) and consult Apify's guide on the legality of web scraping.

Can I integrate Pharma, FDA & Clinical Trials MCP Server with Clay, n8n, Make, Zapier, or Google Sheets?

Yes. Every run outputs clean, flat-friendly fields alongside nested detail objects and integrates natively with Clay, n8n, Make, Zapier, Slack, HubSpot, Airtable, Google Sheets, Airbyte, LangChain, and CrewAI via Apify Integrations. You can also attach webhooks to trigger downstream workflows automatically the instant a run succeeds.

Can I call Pharma, FDA & Clinical Trials MCP Server programmatically via the Apify API?

Yes. Use the official apify-client Python package or Node.js NPM package, or call the synchronous /run-sync-get-dataset-items REST endpoint directly. Full specifications are in the Apify API v2 reference.

Pharma, FDA & Clinical Trials MCP Server not working or need a custom feature?

We actively maintain and monitor all 26 Actors and MCP Servers in our portfolio. If you encounter an edge case, unexpected output, or want to request a new field, please open a ticket on the Actor's Issues tab with your Run ID so we can resolve it rapidly.

來源:README.md,提交 302eda2

工具

0
該伺服器需要 API 金鑰等憑證,因此 SourceWeft 無法在沒有憑證的情況下讀取其工具,也未收錄。

版本歷史

1
  1. v1.0.0最新Nov 19, 2026

評分

已滿足適用檢查項的 85%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具15 分

    請讓伺服器在無憑證時回應 tools/list,或發布可在沙箱中無憑證啟動的軟體包。

  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

其他 Clinical 整合