sql-queries
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted data including uploaded SQL schema files, database documentation, and user-provided diagram descriptions (SKILL.md). While this creates an attack surface for indirect prompt injection, it is the intended primary purpose of the skill and does not present a high risk due to the lack of dangerous capabilities defined in the skill configuration.
- Ingestion points: User-uploaded database_schema.sql files, external documentation text, and schema diagram descriptions.
- Boundary markers: None identified.
- Capability inventory: The skill generates SQL query text and test scripts; it does not have access to shell execution, file-system writing, or network tools in its defined scope.
- Sanitization: None identified.
Audit Metadata