The Agent Skills Directory

[DATA_EXPOSURE]: The utility script scripts/openfda_query.py loads environment variables from the user's ~/.env file to retrieve the FDA_API_KEY. This is a standard and recommended practice for managing API credentials and is performed here to authenticate with the official FDA service.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from the external openFDA API, which represents a potential attack surface for indirect prompt injection if the data source were compromised.
Ingestion points: Data is fetched from https://api.fda.gov using the polite-http library.
Boundary markers: Output is written to JSON files as specified by the --output flag, providing structure but without explicit instructions to the agent to disregard embedded text as potential commands.
Capability inventory: The skill performs network GET requests to the openFDA API and writes results to the local filesystem.
Sanitization: The script performs standard JSON serialization but does not specifically sanitize the internal text content of FDA reports for adversarial patterns.
[EXTERNAL_DOWNLOADS]: The skill's Python script relies on polite-http (a rate-limiting HTTP client) and python-dotenv (for environment variable management). These are standard libraries appropriate for the skill's functionality.
[COMMAND_EXECUTION]: The skill requires execution of a local Python script (scripts/openfda_query.py) using uv to perform its primary functions.

openfda-database