Index Advisor
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and analyze external SQL queries and EXPLAIN plan outputs, which provides a potential vector for instructions embedded within the data.
- Ingestion points: The workflow explicitly asks the agent to collect 'slow query text' and 'EXPLAIN' outputs from the user or database (SKILL.md).
- Boundary markers: The skill does not provide specific delimiters or instructions for the agent to distinguish between the SQL query data and potential instructions embedded in comments.
- Capability inventory: The skill is designed to generate structural database commands (CREATE INDEX, DROP INDEX) which are then potentially executed by the agent or user.
- Sanitization: There are no instructions provided to validate or sanitize the input SQL strings for malicious prompt content before analysis.
Audit Metadata