clickhouse-monitoring

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill is designed to access and display data from sensitive ClickHouse system tables, including system.audit_log, system.sessions, system.login_attempts, and system.text_log. This exposure is fundamental to the skill's primary function as a monitoring and observability tool.
  • [COMMAND_EXECUTION]: The API includes functionality for administrative actions, specifically the ability to terminate running database queries via the kill-query action through the action API endpoint.
  • [PROMPT_INJECTION]: The skill involves a surface for indirect injection due to its dynamic SQL construction pattern. Ingestion points: User-supplied parameters such as lastHours, interval, and table filters are interpolated into SQL query strings. Boundary markers: Documentation examples do not explicitly define delimiters for these inputs. Capability inventory: The skill possesses the ability to read sensitive server logs and execute process termination commands. Sanitization: Standard architectural examples in the development guide show direct string interpolation of parameters without explicit sanitization logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 01:36 AM
Security Audit — agent-trust-hub — clickhouse-monitoring