rabbitmq-expert
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The technical instructions and code examples provided for RabbitMQ management are consistent with professional development standards and prioritize security, including detailed guidance on credential isolation and TLS configuration.
- [INDIRECT_PROMPT_INJECTION]: The skill implements logic to consume and process data from external RabbitMQ message queues, creating an inherent ingestion surface for untrusted data.
- Ingestion points: Data enters the system context through message bodies processed by the
OrderConsumer.process_messagefunction in theapp/consumers.pyexample. - Boundary markers: The instructions lack specific prompt delimiters to separate message data from system instructions, although they recommend strong application-level validation.
- Capability inventory: The skill provides examples for administrative command execution (
rabbitmqctl) and automated testing (pytest). - Sanitization: The documentation explicitly maps security requirements to OWASP standards and emphasizes the importance of validating and sanitizing message content.
Audit Metadata