supabase-audit-authenticated
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Supabase tables, creating a surface for indirect prompt injection from the database content.
- Ingestion points: Data returned from GET requests to tables like users and profiles.
- Boundary markers: None present to distinguish untrusted data from instructions.
- Capability inventory: The agent can execute curl commands and write to local log/evidence files.
- Sanitization: No sanitization of table data is performed before analysis.
- [COMMAND_EXECUTION]: The skill utilizes shell commands (curl, jq) to interact with the Supabase API and parse responses, which is a necessary but powerful capability that could be exploited if inputs are not handled carefully.
Audit Metadata