mental-health-analyzer

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill demonstrates safe operational behavior by limiting its actions to local file reading and writing. It includes extensive disclaimers noting that it is not a medical professional and cannot provide diagnoses or prescriptions. No unauthorized network operations, hardcoded credentials, or obfuscated payloads were identified.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests untrusted text from user-controlled mood diaries.
  • Ingestion points: The skill reads daily mood logs from paths such as data-example/mental-health-logs/YYYY-MM/YYYY-MM-DD.json.
  • Boundary markers: Instructions do not specify delimiters or techniques to isolate embedded instructions in the ingested logs.
  • Capability inventory: The skill is restricted to Read, Grep, Glob, Write, and Edit tools. It lacks the ability to execute shell commands or access the network, which significantly mitigates the potential impact of an injection attack.
  • Sanitization: There is no evidence of input validation or content sanitization for the processed logs.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 05:31 AM