supabase-schema
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a robust safety protocol requiring explicit user confirmation before any database migrations are executed via MCP tools.
- [COMMAND_EXECUTION]: The skill uses database-specific tools to generate types and apply migrations, which are appropriate for its functional scope.
- [PROMPT_INJECTION]: The skill processes user-supplied descriptions to generate executable SQL code. This indirect prompt injection surface is documented; however, the risk is mitigated by the mandatory review process and the context of a database design tool.
- Ingestion points: User descriptions of entities and relationships (SKILL.md).
- Boundary markers: None.
- Capability inventory: apply_migration, generate_typescript_types (SKILL.md).
- Sanitization: Human review and confirmation (Step 7).
Audit Metadata