sleep-analyzer

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its architecture for processing external health data. \n
  • Ingestion points: The skill ingests data from multiple potentially untrusted sources, including 'sleep-tracker.json', 'fitness-tracker.json', 'diet-records/', and 'mood-tracker.json' via the Read, Grep, and Glob tools. \n
  • Boundary markers: The instructions lack explicit boundary markers or delimiters that would inform the agent to ignore any natural language instructions embedded within the health records. \n
  • Capability inventory: The skill is granted 'Write' access to the file system, which provides a functional capability that could be exploited if an injection attack were successful. \n
  • Sanitization: There is no description of data validation or sanitization logic to verify the integrity of the records before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 05:50 PM