finding-expensive-queries
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze data from Snowflake's
QUERY_HISTORY, which includes thequery_textfield. This field contains arbitrary SQL statements that could potentially contain malicious instructions aimed at influencing the agent's behavior. - Ingestion points: Data is retrieved from the
SNOWFLAKE.ACCOUNT_USAGE.QUERY_HISTORYview inSKILL.md. - Boundary markers: The instructions do not define specific delimiters to separate the retrieved query text from other instructions, nor do they include warnings to ignore instructions embedded in the data.
- Capability inventory: The skill allows the agent to read query performance metadata; actual execution risks would depend on the availability of other tools (e.g., shell or write access) in the agent's environment.
- Sanitization: The skill does not provide methods for sanitizing or filtering the query text before presenting it to the agent for analysis.
Audit Metadata