supabase-sql
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied SQL migrations, which presents a surface for indirect prompt injection if the input contains instructions designed to override the agent's behavior.
- Ingestion points: User-provided SQL code passed to the agent for formatting (referenced throughout SKILL.md).
- Boundary markers: The skill does not implement specific delimiters or safety instructions to isolate the input data from the prompt context.
- Capability inventory: The skill instructions in SKILL.md direct the agent to write telemetry data to ~/.claude/skill-analytics/last-outcome-supabase-sql.json and generate SQL code output.
- Sanitization: No sanitization or validation is applied to the input SQL strings.
Audit Metadata