tl-pg-boss

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for background job processing where workers ingest data stored in a PostgreSQL database, creating a potential surface for indirect prompt injection if payloads are not handled securely by the implementing application.
  • Ingestion points: Job handlers receive untrusted payloads via the job.data property in SKILL.md and references/typescript-patterns.md.
  • Boundary markers: No explicit natural language delimiters or 'ignore' instructions are provided for job data payloads in the example code.
  • Capability inventory: The documented handlers perform various tasks including sending emails, processing images, and executing database queries.
  • Sanitization: The skill recommends using idempotent handlers and upserts as a reliability best practice, though it does not provide specific data sanitization examples.
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing well-known Node.js packages including pg-boss, @pg-boss/dashboard, and the third-party pg-boss-admin-dashboard utility via standard package managers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:40 PM