query-writing

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill possesses a surface for indirect prompt injection because it processes data from external database sources.
  • Ingestion points: Data enters the agent context via sql_db_schema and sql_db_query results.
  • Boundary markers: None. The instructions do not specify using delimiters or warnings to ignore instructions found within database records or schema names.
  • Capability inventory: The skill is capable of executing arbitrary SELECT queries through sql_db_query.
  • Sanitization: No sanitization or validation logic is defined to check for malicious content within the database outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:20 PM