Skills fda-database
💊

fda-database

Safe ⚡ Contains scripts🌐 Network access🔑 Env variables

Query FDA drug and device databases

Also available from: davila7

Access comprehensive FDA regulatory data including drug adverse events, medical device recalls, and safety information through the official openFDA API. Build pharmacovigilance and regulatory research workflows with Python.

Supports: Claude Codex Code(CC)
🥈 77 Silver
1

Download the skill ZIP

2

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

3

Toggle on and start using

Test it

Using "fda-database". Find recalls for metformin diabetes medications

Expected outcome:

  • Metformin recalls: 3 recent recalls found
  • Reason: NDMA impurity detection
  • Classification: Class II
  • Status: Ongoing

Using "fda-database". Show device adverse events for pacemakers

Expected outcome:

  • Total adverse events: 15,432
  • Event types: Death (234), Injury (4,521), Malfunction (10,677)
  • Most common issue: Battery depletion

Using "fda-database". Look up UNII code for ibuprofen

Expected outcome:

  • UNII: Y43UW15LVD
  • CAS Number: 15687-27-1
  • Molecular Formula: C13H18O2
  • Substance Class: chemical

Security Audit

Safe
v4 • 1/17/2026

This is a legitimate API wrapper for the FDA's public openFDA API (api.fda.gov). All static findings are false positives: the 'CRITICAL' heuristic flagged code execution + network + credential access as suspicious, but this is standard API client behavior for authenticating with a legitimate government API. MD5 is used for cache key generation, not cryptographic security. Documentation code blocks were misidentified as shell commands. All network requests go to the official FDA API endpoint with no exfiltration or credential theft patterns.

11
Files scanned
4,787
Lines analyzed
3
findings
4
Total audits

Risk Factors

⚡ Contains scripts (2)
scripts/fda_query.py:1-441 scripts/fda_examples.py:1-336
🌐 Network access (2)
scripts/fda_query.py:87 scripts/fda_query.py:135
🔑 Env variables (2)
scripts/fda_query.py:100 scripts/fda_examples.py:299
Audited by: claude View Audit History →

Quality Score

68
Architecture
100
Maintainability
85
Content
31
Community
100
Security
78
Spec Compliance

What You Can Build

Drug safety monitoring

Monitor adverse event reports and identify safety signals for medications

Device surveillance

Track medical device adverse events and recall status for products

Regulatory compliance

Research approval pathways and compliance history for submissions

Try These Prompts

Basic drug lookup 
Use the FDA database skill to look up adverse events for aspirin and show the 5 most common reactions
Device recall search 
Query the FDA database for medical device recalls involving pacemakers in the past year
Comparative analysis 
Compare adverse event rates between ibuprofen, naproxen, and aspirin using the FDA database
Food safety monitoring 
Monitor food recalls related to undeclared peanuts and show the most recent 10 recalls

Best Practices

  • Set the FDA_API_KEY environment variable to increase rate limits from 1,000 to 120,000 requests per day
  • Use specific search terms and field queries for accurate results rather than broad wildcards
  • Implement caching for frequently accessed data to reduce API calls and improve performance

Avoid

  • Do not use overly broad wildcard searches that return excessive unrelated results
  • Avoid making rapid successive requests without rate limiting, which may trigger 429 errors
  • Do not assume all records contain all fields; always check field existence before accessing

Frequently Asked Questions

Is an API key required?
No, the FDA API works without a key but has lower rate limits (1,000/day vs 120,000/day with a key).
What data categories are available?
Drugs, devices, foods, animal/veterinary products, and chemical substances are supported.
How current is the data?
Data is updated regularly but may have a lag of days to weeks depending on the endpoint.
Can I download full datasets?
The API returns paginated results. For bulk data, consider downloading official FDA data files separately.
What are the rate limits?
240 requests per minute with or without a key. Daily limits are 1,000 without a key and 120,000 with a key.
Is this data suitable for clinical decisions?
This is public post-market surveillance data. Use official prescribing information for clinical decisions.

Developer Details

Author

K-Dense-AI

License

MIT

Repository

https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/fda-database

Ref

main

File structure

📁 references/

📄 animal_veterinary.md

📄 api_basics.md

📄 devices.md

📄 drugs.md

📄 foods.md

📄 other.md

📁 scripts/

📄 fda_examples.py

📄 fda_query.py

📄 evaluation.json

📄 SKILL.md