sink-clickhouse-jsonl

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill validates ClickHouse identifiers (database and table names) using a strict whitelist regular expression (^[A-Za-z_][A-Za-z0-9_]{0,254}$) to prevent SQL injection in identifier positions.
  • [SAFE]: Credentials and connection settings are managed through standard environment variables (CLICKHOUSE_HOST, CLICKHOUSE_USER, CLICKHOUSE_PASSWORD), which is the recommended practice for secure secret handling.
  • [SAFE]: The skill implements a safety-first approach by defaulting to a dry-run mode, requiring an explicit --apply flag for actual database write operations.
  • [SAFE]: Network egress is restricted to the official *.clickhouse.cloud domain, aligning with the skill's purpose and reducing the risk of unauthorized data exfiltration.
  • [SAFE]: The implementation uses the official clickhouse-connect driver and avoids dangerous operations like eval(), exec(), or arbitrary SQL execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 01:17 AM
Security Audit — agent-trust-hub — sink-clickhouse-jsonl