drug-interaction-checker
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted patient medication regimens.\n
- Ingestion points: User-supplied medication lists and drug names (SKILL.md).\n
- Boundary markers: Absent; the documentation provides no delimiters or instructions to ignore embedded commands within user input.\n
- Capability inventory: None identified; the skill contains no scripts or tools, but provides clinical assessments based on input (SKILL.md).\n
- Sanitization: Absent; no input validation is defined to filter malicious text within the medication data.\n- [EXTERNAL_DOWNLOADS]: The skill installation instructions fetch content from the author's own repository.\n
- Source: https://github.com/Open-Medica/open-medical-skills\n
- Method: npx skills add command.\n
- Note: The source repository belongs to the skill author (open-medica).
Audit Metadata