hunt-supabase

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs discovering, extracting, and embedding Supabase anon keys and JWTs into curl commands and scripts (e.g., -H "Authorization: Bearer $ANON_KEY"), which requires the LLM/agent to handle and output secret values verbatim, creating an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive guide: it contains step-by-step instructions to discover Supabase project credentials, enumerate and exfiltrate database tables, obtain JWTs, bypass RLS (cross-tenant access), enumerate users, and list/download/upload storage — all facilitating unauthorized access and data theft.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes a Supabase exploitation workflow that ingests free text from outsider-controlled target web content at runtime (e.g., downloading JS bundles from https://$TARGET and extracting configs from /env, source maps, and then searching them for Supabase URL/anon key strings).