supabase-keys
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill demonstrates good security practices by instructing the agent to never echo sensitive API keys back to the user and to explicitly reject high-privilege service_role keys.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data (Supabase URL and API key) provided by the user to populate local environment files.
- Ingestion points: The agent is instructed to ask the user for a project URL and an anon key in the chat.
- Boundary markers: There are no explicit boundary markers or delimiters defined for the values being pasted into the file instructions.
- Capability inventory: The agent is granted the capability to write to the local environment file (.env.local).
- Sanitization: The skill includes a validation step to ensure a service_role key is not used by mistake, but it does not perform general sanitization on the input strings.
Audit Metadata