OpenFDA MCP Server
MCP server providing access to U.S. FDA public datasets including drugs, medical devices, adverse events, recalls, and more through standardized tools.
README
OpenFDA MCP Server
An MCP (Model Context Protocol) server that provides access to U.S. FDA public datasets including drugs, medical devices, adverse events, recalls, and more.
Features
- 10 FDA Tools: Query drug and device databases through a standardized MCP interface
- Drug Tools: Adverse events (FAERS), labels, NDC directory, recalls, Drugs@FDA, shortages
- Device Tools: 510(k) clearances, classifications, adverse events (MDR), recalls
- Individual Record Lookup: Retrieve specific reports by ID (safety report ID, set ID, MDR report number)
- Advanced Filtering: Filter drug labels by boxed warnings, request specific label sections
- Rate Limit Aware: Supports authenticated requests for higher rate limits (120k/hour vs 1k/hour)
- TypeScript: Fully typed with Zod schema validation
Installation
# Install globally
npm install -g openfda-mcp-server
# Or use with npx
npx openfda-mcp-server
Configuration
Claude Desktop
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"openfda": {
"command": "npx",
"args": ["-y", "openfda-mcp-server"],
"env": {
"OPENFDA_API_KEY": "your-api-key-optional"
}
}
}
}
Environment Variables
| Variable | Description | Default |
|---|---|---|
OPENFDA_API_KEY |
FDA API key for higher rate limits | None (uses public limits) |
Get a free API key at: https://open.fda.gov/apis/authentication/
Available Tools
Drug Tools
| Tool | Description |
|---|---|
search_drug_adverse_events |
Search FAERS for drug safety reports. Retrieve specific report by safetyReportId. |
search_drug_labels |
Search drug labeling/prescribing info. Get by setId, filter by hasBoxedWarning, select sections. |
search_drug_ndc |
Search the National Drug Code directory |
search_drug_recalls |
Search drug recall enforcement reports |
search_drugs_at_fda |
Search approved drug applications |
search_drug_shortages |
Search current and resolved drug shortages |
Device Tools
| Tool | Description |
|---|---|
search_device_510k |
Search 510(k) premarket notifications |
search_device_classifications |
Search device classification database |
search_device_adverse_events |
Search MDR adverse event reports. Retrieve specific report by reportNumber. |
search_device_recalls |
Search device recall enforcement reports |
Tool Parameters
search_drug_labels
| Parameter | Type | Description |
|---|---|---|
setId |
string | Retrieve a specific label by its unique set_id |
drugName |
string | Drug brand or generic name |
indication |
string | Medical indication or use case |
activeIngredient |
string | Active ingredient/substance name |
route |
string | Route of administration (e.g., 'oral', 'intravenous') |
hasBoxedWarning |
boolean | Filter for drugs with boxed warnings (most serious warnings) |
sections |
string[] | Specific label sections to return (see below) |
limit |
number | Maximum results (1-100) |
skip |
number | Number of results to skip for pagination |
Available sections: indications_and_usage, dosage_and_administration, contraindications, warnings, warnings_and_cautions, adverse_reactions, drug_interactions, clinical_pharmacology, mechanism_of_action, pharmacokinetics, overdosage, description, how_supplied, storage_and_handling, boxed_warning
search_drug_adverse_events
| Parameter | Type | Description |
|---|---|---|
safetyReportId |
string | Retrieve a specific FAERS report by its 8-digit ID |
drugName |
string | Drug or product name to search |
reaction |
string | Adverse reaction (e.g., 'headache', 'nausea') |
manufacturer |
string | Drug manufacturer name |
serious |
boolean | Filter for serious adverse events only |
dateFrom |
string | Start date (YYYY-MM-DD) |
dateTo |
string | End date (YYYY-MM-DD) |
limit |
number | Maximum results (1-100) |
skip |
number | Number of results to skip for pagination |
search_device_adverse_events
| Parameter | Type | Description |
|---|---|---|
reportNumber |
string | Retrieve a specific MDR report by report number |
deviceName |
string | Device generic name |
brandName |
string | Device brand name |
manufacturerName |
string | Device manufacturer name |
eventType |
string | Type: 'Injury', 'Malfunction', 'Death', 'Other' |
dateFrom |
string | Start date (YYYY-MM-DD) |
dateTo |
string | End date (YYYY-MM-DD) |
limit |
number | Maximum results (1-100) |
skip |
number | Number of results to skip for pagination |
Usage Examples
Search for drug adverse events
Find adverse events for aspirin in the last year that were serious
Get a specific adverse event report
Retrieve FAERS report with safety report ID 10003641
Search for drugs with boxed warnings
Find all drugs with boxed warnings related to cardiovascular risks
Get specific label sections
Get the boxed_warning and contraindications sections for warfarin
Search for drug recalls
Search for Class I drug recalls from Pfizer
Search for device clearances
Find 510(k) clearances for cardiac pacemakers
Get a specific device adverse event report
Retrieve MDR report number 2649622-2020-08294
Search for device adverse events
Search device adverse events for insulin pumps that resulted in injury
CLI Usage
# Start with stdio transport (default)
openfda-mcp-server
# Start with HTTP transport
openfda-mcp-server --transport http --port 3000
Docker
Quick Start
# Build and run with docker-compose
docker-compose up -d
# Or build manually
docker build -t openfda-mcp-server .
docker run -p 3000:3000 -e OPENFDA_API_KEY=your-key openfda-mcp-server
Docker Compose
services:
openfda-mcp:
image: openfda-mcp-server:latest
ports:
- "3000:3000"
environment:
- OPENFDA_API_KEY=${OPENFDA_API_KEY:-}
Environment Variables
Pass environment variables to the container:
docker run -p 3000:3000 \
-e OPENFDA_API_KEY=your-api-key \
-e DEBUG=openfda-mcp:* \
openfda-mcp-server
Development
# Install dependencies
pnpm install
# Run in development mode
pnpm dev
# Test locally
pnpm serve:test
# Run full validation (format, lint, test, build)
pnpm validate
# Build for production
pnpm build
Programmatic Usage
import { createOpenFDAServer, handleSearchDrugAdverseEvents, handleSearchDrugLabels } from "openfda-mcp-server/lib"
// Create a custom server
const server = createOpenFDAServer({
name: "my-fda-server",
version: "1.0.0",
})
// Search for adverse events
const adverseEvents = await handleSearchDrugAdverseEvents({
drugName: "aspirin",
serious: true,
limit: 10,
})
// Get a specific adverse event by ID
const specificReport = await handleSearchDrugAdverseEvents({
safetyReportId: "10003641",
})
// Search for drugs with boxed warnings
const boxedWarningDrugs = await handleSearchDrugLabels({
hasBoxedWarning: true,
limit: 10,
})
// Get specific sections of a drug label
const warfarinLabel = await handleSearchDrugLabels({
setId: "0cbce382-9c88-4f58-ae0f-532a841e8f95",
sections: ["boxed_warning", "contraindications", "adverse_reactions"],
})
API Rate Limits
| Authentication | Per Minute | Per Hour |
|---|---|---|
| Without API Key | 40 | 1,000 |
| With API Key | 240 | 120,000 |
License
MIT
Acknowledgments
- OpenFDA - FDA open data API
Sponsored by <a href="https://sapientsai.com/"><img src="https://sapientsai.com/images/logo.svg" alt="SapientsAI" width="20" style="vertical-align: middle;"> SapientsAI</a> — Building agentic AI for businesses
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.