reputation-sentiment

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection.
  • Ingestion points: Reads review text from external websites via WebSearch and WebFetch tools (SKILL.md).
  • Boundary markers: None identified.
  • Capability inventory: Performs file writes to the local directory to save analysis reports.
  • Sanitization: No explicit sanitization or filtering of review content is defined before processing or report generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 11:40 PM
Security Audit — agent-trust-hub — reputation-sentiment