supabase
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process content from external URLs (e.g., https://supabase.com/llms.txt) using a sub-agent. Ingesting external data into the agent's context is an inherent surface for indirect prompt injection, though the use of official, well-known service domains for documentation research is a standard and expected practice.\n
- Ingestion points: Documentation files fetched from the supabase.com domain via WebFetch tools (SKILL.md).\n
- Boundary markers: The research prompt template does not explicitly define delimiters or instructions to ignore potential injection within the fetched content.\n
- Capability inventory: The skill is capable of spawning sub-agents and performing network fetches to gather information.\n
- Sanitization: There is no mention of sanitizing, filtering, or escaping the retrieved documentation content before processing.
Audit Metadata