supabase

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches documentation and changelogs from official Supabase domains (supabase.com) and GitHub repositories (github.com/supabase), which are well-known and trusted sources.
  • [COMMAND_EXECUTION]: Provides instructions for standard Supabase CLI usage (e.g., supabase db query, supabase migration new) and network connectivity tests (curl to Supabase MCP server). These are documented for legitimate administrative and troubleshooting purposes.
  • [DATA_EXFILTRATION]: No exfiltration patterns detected. The skill contains proactive instructions to prevent data exposure, such as warnings against putting sensitive data in user_metadata or exposing service_role keys.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes instructions to process external data (changelogs and documentation). It provides standard security checklists to mitigate common configuration errors, although the agent should always treat documentation content as data.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials found. The skill correctly instructs users to manage secrets via environment variables and warns against exposing private keys in frontend code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 12:45 AM
Security Audit — agent-trust-hub — supabase