clickhouse-best-practices
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a comprehensive resource for ClickHouse database management, consisting of 28 specific rules for optimization and 3 rules for agent safety. All content aligns with the stated purpose of assisting developers with ClickHouse.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing user-provided SQL queries and table schemas. Although this creates an ingestion point for untrusted data, the skill includes critical safety rules (e.g.,
rules/agent-query-safety.md) that mandate the use ofLIMIT,max_rows_to_read, andmax_execution_timeto prevent resource exhaustion or broad data access. - [EXTERNAL_DOWNLOADS]: The skill references standard vendor-provided tools, including the
mcp-clickhousePython package and ClickHouse Cloud connection endpoints (*.clickhouse.cloud). These are official resources from ClickHouse Inc. and are treated as safe. - [COMMAND_EXECUTION]: The documentation provides examples of standard CLI usage and connectivity tests using
curlandclickhouse-client. These commands are standard administrative procedures for the target technology and do not involve suspicious execution patterns.
Audit Metadata