optimizing-sql-queries
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of SQL queries and execution plan outputs which could contain malicious instructions designed to influence agent behavior.
- Ingestion points: The skill requires the user to provide slow SQL query text and
EXPLAINoutput (referenced in SKILL.md instructions). - Boundary markers: The instructions do not define clear delimiters or specific guidance for the agent to ignore instructions that might be embedded within the SQL comments or data.
- Capability inventory: The skill frontmatter allows the use of
Bashtools scoped for database clients (psql,mysql,mongosh), providing a path for potential impact if a prompt injection is successful. - Sanitization: There are no validation or sanitization steps described to ensure the input SQL or execution plans are safe to process.
Audit Metadata