supabase

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to fetch and scan public web content at runtime (e.g., https://supabase.com/changelog.md and linked docs pages), which are outsider-authored free text that can be ingested into the LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to fetch external markdown docs that steer its behavior (e.g., https://supabase.com/changelog.md and https://supabase.com/docs/guides/security/product-security.md), meaning remote content will directly control prompts and guidance.