medical
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it ingests untrusted user data, such as medication names and symptom descriptions, which are then stored and later retrieved to generate summaries.
- Ingestion points: User-provided health data is ingested via commands and stored in the
memory/health/directory. - Boundary markers: The documentation lacks specific delimiters or instructions for the agent to ignore potentially malicious instructions embedded within the stored medical data.
- Capability inventory: The skill leverages several Python scripts in the
scripts/directory to read and write files and generate summaries. - Sanitization: There is no evidence of sanitization or input validation to prevent user-supplied text from influencing subsequent agent actions during summary generation.
- [COMMAND_EXECUTION]: The skill is designed to execute local Python scripts (e.g.,
add_medication.py,add_vital.py) to manage its data operations. While these are local and functional, the lack of input sanitization at the script interface remains a potential concern.
Audit Metadata