bull-bullmq-nestjs

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements a task processing architecture that is susceptible to indirect prompt injection if untrusted data is processed.
  • Ingestion points: The InternalJobsController in references/cloud-tasks-migration.md exposes a POST endpoint (/internal/jobs/:queue/:jobName) designed to receive task payloads from external sources (GCP Cloud Tasks).
  • Boundary markers: The provided code examples lack explicit delimiters or instructions (e.g., 'ignore embedded commands') to prevent the agent or worker from interpreting data within the job payload as instructions.
  • Capability inventory: The NotificationsProcessor and RemindersProcessor have the capability to perform network operations (sending Firebase push notifications) and execute database queries via Prisma based on the content of the ingested jobs.
  • Sanitization: There is no evidence of content sanitization or validation for the message bodies (title, body, deepLink) before they are sent to the push notification service.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 05:58 PM