The Agent Skills Directory

[SAFE]: Comprehensive analysis of the skill's instructions and scripts reveals no evidence of malicious intent. The skill operates as described, providing structured access to public FDA datasets.
[CREDENTIALS_UNSAFE]: The skill manages authentication safely by directing users to obtain their own API keys from the official openFDA website. It utilizes environment variables for key management and correctly uses placeholders (e.g., 'YOUR_API_KEY') in examples to avoid credential exposure.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to the official government domain api.fda.gov to retrieve regulatory information. This communication is restricted to the intended purpose of the skill and targets a well-known, trusted service.
[COMMAND_EXECUTION]: The Python scripts (scripts/fda_query.py and scripts/fda_examples.py) use the standard requests library for API communication. There is no evidence of unsafe subprocess calls, shell command execution, or dynamic code evaluation risks.

fda-database