clickhouse-performance-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes queries from the
system.query_logtable, which can include arbitrary text from user-executed queries. \n - Ingestion points: Database query text is ingested via
system.query_loginSKILL.mdandreferences/implementation.md. \n - Boundary markers: No specific boundary markers or instructions are included to prevent the agent from executing instructions that might be embedded in the query logs. \n
- Capability inventory: The agent has access to
Read,Write, andEdittools and is instructed to perform database modifications (e.g.,ALTER TABLE,CREATE TABLE). \n - Sanitization: The skill lacks sanitization logic to filter or escape instructions potentially hidden in the query history data.
Audit Metadata