etl-sync-job-builder

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides examples of ETL jobs that ingest external data and process it using raw SQL queries. Specifically, the OrdersETLJob example in SKILL.md demonstrates manual SQL query construction using string interpolation (${batch.map(...).join(",")}) rather than parameterized queries. This creates a vulnerability surface where malicious data from a source system could perform SQL injection if executed as written.
  • Ingestion points: The extract method in SKILL.md fetches data from source databases or APIs.
  • Boundary markers: No boundary markers or instructions to sanitize source data are provided in the implementation examples.
  • Capability inventory: The code utilizes prisma.$executeRaw to perform write operations to a destination database.
  • Sanitization: The example logic lacks proper parameterization for batch inserts, relying on template literal interpolation which fails to escape potentially malicious input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:30 AM
Security Audit — agent-trust-hub — etl-sync-job-builder