The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill makes requests to https://dailymed.nlm.nih.gov, which is the official National Library of Medicine (NLM) REST API for FDA drug labels. This is a trusted government resource.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from the DailyMed API, which serves as a potential ingestion surface for indirect prompt injection.
Ingestion points: API responses are ingested via requests.get().json() calls in SKILL.md across various queries.
Boundary markers: No specific boundary markers or instructions to ignore embedded content are present; responses are handled as standard JSON objects.
Capability inventory: The skill is limited to performing HTTP GET requests and generating local data files (CSV, PNG). There are no capabilities for command execution, dynamic evaluation, or unauthorized file system modification.
Sanitization: Uses standard JSON parsing; no additional text-based sanitization is implemented, though the risk is mitigated by the trusted nature of the source and limited capabilities.
[SAFE]: No malicious patterns such as obfuscation, hardcoded credentials, or persistence mechanisms were detected. The skill demonstrates best practices by implementing rate limiting (time.sleep(0.3)) and using standardized drug identifiers (RxCUI and NDC).

dailymed-database