notis-reports

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines clear boundaries and utilizes platform-specific tools for report generation without introducing high-risk patterns or external code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external information to populate reports, which creates a potential surface for indirect injection.
  • Ingestion points: The skill gathers information from user context or external data to populate report sections and KPIs (SKILL.md).
  • Boundary markers: Instructions specify using native document links and revision-based conflict resolution to maintain state integrity (SKILL.md).
  • Capability inventory: The skill calls LOCAL_NOTIS_SAVE_REPORT and LOCAL_NOTIS_SAVE_HTML_DOCUMENT to persist data (SKILL.md).
  • Sanitization: The documentation explicitly states that HTML reports are isolated from the Portal session and interactive components remain within a sandbox (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 05:34 PM
Security Audit — agent-trust-hub — notis-reports