setup-rls

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): The skill is purely informational and instructional. It guides the AI agent on how to help users configure database security policies.
  • [PROMPT_INJECTION] (SAFE): No evidence of prompt injection, role-play bypasses, or instructions to ignore safety filters were found in the markdown or YAML frontmatter.
  • [DATA_EXFILTRATION] (SAFE): There are no hardcoded credentials, sensitive file paths, or network-bound commands. The SQL examples use standard Supabase authentication functions (auth.uid(), auth.jwt()).
  • [REMOTE_CODE_EXECUTION] (SAFE): The skill does not include any commands for downloading or executing remote scripts. No package managers (npm, pip) or shell pipes are used.
  • [INDIRECT_PROMPT_INJECTION] (LOW): As a skill that processes user requests regarding security and permissions, it inherently has a surface for indirect prompt injection if a user provides malicious input. However, the skill does not grant elevated system privileges and its capabilities are limited to local file editing tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 02:18 AM