clickhouse-io

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The TypeScript code examples for data insertion (bulkInsertTrades and insertTrade) construct SQL queries using string interpolation with variable values (e.g., '${trade.id}'). This is a common vulnerability pattern that can lead to SQL injection if the input values are not properly sanitized or if parameterized queries are not used.
  • [PROMPT_INJECTION]: The skill describes patterns for ingesting data from external sources, such as PostgreSQL and Change Data Capture (CDC) streams. This introduces a surface for indirect prompt injection where malicious content in the source data could influence the agent's behavior if it later processes that data. 1. Ingestion points: PostgreSQL database records and CDC notification payloads (msg.payload). 2. Boundary markers: None specified in the instructional patterns. 3. Capability inventory: The skill demonstrates database write operations via clickhouse.query and clickhouse.insert. 4. Sanitization: The examples do not include input validation or sanitization logic for the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 07:42 AM
Security Audit — agent-trust-hub — clickhouse-io