medication

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it transforms natural language user input into structured data for local storage without using safety delimiters or input validation.\n
  • Ingestion points: The medication_description_natural_language argument in SKILL.md is the primary entry point for untrusted user data.\n
  • Boundary markers: Absent. There are no instructions for the agent to use delimiters or to ignore potential instructions embedded within the medication description.\n
  • Capability inventory: The skill uses the Write tool to save records to data/medications/medications.json and logs to data/medication-logs/. This capability could be exploited if an injection causes the agent to write malicious or incorrect data.\n
  • Sanitization: Absent. The skill does not specify any validation, escaping, or filtering mechanisms for the data extracted from user input before it is written to the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 05:26 AM
Security Audit — agent-trust-hub — medication