looker-studio-report

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): No security issues were detected. The skill provides a structured workflow and template for manual report design and documentation.
  • Indirect Prompt Injection (LOW): The skill ingests user-provided project details and business context to populate templates. This is a low-risk surface as the output is purely informational markdown and does not trigger automated system actions.
  • Ingestion points: Step 2 in SKILL.md asks for project/dataset context and business background.
  • Boundary markers: None explicitly defined for user input interpolation.
  • Capability inventory: No subprocess calls, network ops, or file writes are performed by scripts; the skill only generates text documentation.
  • Sanitization: Not applicable as no automated execution occurs.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:34 PM