The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from external sources like PostgreSQL and ClickHouse system logs without sanitization or boundary markers. * Ingestion points: extractFromPostgres and pgClient.on('notification') in SKILL.md. * Boundary markers: Absent. * Capability inventory: Database write operations via clickhouse.query and clickhouse.insert in SKILL.md. * Sanitization: Absent in the provided logic.
[COMMAND_EXECUTION]: TypeScript code snippets demonstrate constructing SQL queries using string interpolation (template literals). This is a known SQL injection vector if the input data is not properly escaped or parameterized.

clickhouse-io