supabase-platform-weapon

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates the deployment of Supabase assets using the official Supabase CLI (supabase db push, supabase functions deploy, supabase link) and curl. These operations are standard for platform management and are executed using personal access tokens.
  • [DATA_EXFILTRATION]: The instructions handle sensitive management credentials, such as the SUPABASE_ACCESS_TOKEN and SUPABASE_SERVICE_ROLE_KEY. It correctly identifies the service_role_key as a high-privilege credential that bypasses RLS and provides guidance to ensure it is never exposed to the client side. Data transmissions are directed exclusively to the official Supabase Management API.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection (Category 8) because it processes external SQL migrations and Edge Function code.
  • Ingestion points: Files located in supabase/migrations/, supabase/functions/, and environment configuration files.
  • Boundary markers: None identified within the provided instruction files.
  • Capability inventory: The skill allows for arbitrary SQL execution via the Supabase Management API and shell command execution via the Supabase CLI.
  • Sanitization: The skill explicitly mandates the use of jq -Rs to safely escape multi-line SQL content into JSON payloads, preventing common injection errors during API calls.
  • [EXTERNAL_DOWNLOADS]: The skill references reputable external resources and services, including the official Supabase documentation, the Supabase CLI, Deno, and the Supabase JavaScript SDK (via JSR). All references are to well-known and trusted organizations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 03:22 AM
Security Audit — agent-trust-hub — supabase-platform-weapon