rabbitmq-rails

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a template for background workers that ingest data from external message queues.
  • Ingestion points: The work method in OrderPlacedWorker (SKILL.md) receives a payload from the message broker.
  • Boundary markers: The worker template does not utilize delimiters or specific instructions to isolate or ignore potentially malicious content within the payload.
  • Capability inventory: The worker logic is capable of performing database operations, such as transactions and record creation (ApplicationRecord.transaction, ProcessedEvent.create!).
  • Sanitization: The payload is parsed via JSON.parse, but no further structure validation or input sanitization is demonstrated in the snippet.
  • [SAFE]: The skill follows secure configuration patterns by fetching connection strings from environment variables (ENV.fetch("RABBITMQ_URL")).
  • [SAFE]: The skill references trusted and well-known sources, including official RabbitMQ documentation and established Ruby gem repositories on GitHub.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:48 AM
Security Audit — agent-trust-hub — rabbitmq-rails