clickhouse-io
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill defines patterns for ingesting and processing data from external sources, which presents a surface for indirect prompt injection. 1. Ingestion points: External data enters the agent context through the ETL and CDC patterns (e.g., via PostgreSQL notifications) in SKILL.md. 2. Boundary markers: No explicit delimiters or instructions are used to separate untrusted data from system instructions. 3. Capability inventory: Includes database query and insertion capabilities using the 'clickhouse' package. 4. Sanitization: Code snippets use string interpolation for SQL construction, which does not provide sanitization against embedded instructions or potential data-driven injection.
- [EXTERNAL_DOWNLOADS]: References standard and well-known Node.js libraries 'clickhouse' and 'pg' for database connectivity.
Audit Metadata