Fda Intelligence Mcp
Pricing
from $50.00 / 1,000 tool calls
Fda Intelligence Mcp
Production-ready MCP server for FDA drug intelligence. Query approvals, recalls, drug labels, and pipeline activity via AI-native tools. Built for pharma competitive intelligence, biotech analysts, and investment researchers.
Pricing
from $50.00 / 1,000 tool calls
Rating
0.0
(0)
Developer
MCP Pro
Actor stats
0
Bookmarked
1
Total users
0
Monthly active users
16 hours ago
Last modified
Categories
Share
A production-ready Model Context Protocol (MCP) server that gives AI agents and LLM applications direct access to the FDA's public drug database. Built for pharma competitive intelligence teams, biotech analysts, medical affairs professionals, and investment researchers.
Overview
FDA Intelligence MCP wraps the OpenFDA public API into four semantically-named tools following the fda_{action}_{resource} convention. Deployed as an Apify Actor using Streamable HTTP transport with pay-per-event billing.
Available Tools
| Tool | Purpose | Key Parameters |
|---|---|---|
fda_get_drug_approvals | Search FDA drug approval records (CDER) | drug_name, applicant, application_type, date_from, date_to |
fda_get_safety_alerts | Retrieve recalls, warnings, and market withdrawals | drug_name, recall_class, date_from, date_to |
fda_get_drug_label | Fetch full FDA-approved prescribing information | drug_name, label_type |
fda_get_pipeline_activity | Analyze pre-approval competitive pipeline activity | applicant, drug_class, application_type, date_from |
All tools return structured JSON with calls_remaining, total_results, returned_count, has_more, and structured error dicts on failure.
Quick Start
Local Development
git clone https://github.com/MCPPro/fda-intelligence-mcp.gitcd fda-intelligence-mcppython -m venv .venvsource .venv/bin/activate # Windows: .venv\Scripts\activatepip install -r requirements.txtcp .env.example .env# Optionally add your OPENFDA_API_KEYcd src && python server.py
Docker
docker build -t fda-intelligence-mcp .docker run -p 8000:8000 -e OPENFDA_API_KEY=your_key fda-intelligence-mcp
Run Tests
$pytest tests/test_server.py -v
Authentication
| Mode | Rate Limit |
|---|---|
| No API key | 40 requests/min, 1,000/day |
| With API key | 240 requests/min, 120,000/day |
Get a free key at open.fda.gov/apis/authentication.
Example Queries
1. Which GLP-1 drugs did Novo Nordisk get approved in 2022–2023?
{ "tool": "fda_get_drug_approvals", "params": { "applicant": "Novo Nordisk", "application_type": "NDA", "date_from": "2022-01-01", "date_to": "2023-12-31" } }
2. Are there any Class I recalls for metformin?
{ "tool": "fda_get_safety_alerts", "params": { "drug_name": "metformin", "recall_class": "I", "date_from": "2022-01-01" } }
3. What are the approved indications and warnings for Keytruda?
{ "tool": "fda_get_drug_label", "params": { "drug_name": "keytruda", "label_type": "prescription" } }
4. What BLA submissions has AstraZeneca filed since 2023?
{ "tool": "fda_get_pipeline_activity", "params": { "applicant": "AstraZeneca", "application_type": "BLA", "date_from": "2023-01-01" } }
Rate Limits and Pricing
The server implements a 1 request/second rate limiter and 1-hour TTL cache. Apify billing is $0.01 per tool call via Actor.charge(event_name='tool-call').
Data Source
All data from the OpenFDA public database (U.S. FDA). Public domain data per U.S. government policy.
- Drug Approvals:
/drug/drugsfda.json - Safety Alerts:
/drug/enforcement.json - Drug Labels:
/drug/label.json