event-driven-architecture
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill documentation includes implementation patterns that ingest data from external sources, creating a surface for indirect prompt injection.
- Ingestion points: In
SKILL.md, Step 4 and Step 8 provide Python code that retrieves and processes messages from Kafka (self.consumer.poll) and RabbitMQ. - Boundary markers: The code examples do not demonstrate the use of delimiters or specific instructions to isolate or ignore potentially malicious instructions embedded within the message payloads.
- Capability inventory: The skill scripts include network operations (broker communication), database execution (PostgreSQL event store and outbox tables), and architectural patterns that could be manipulated if an agent processes a malicious event.
- Sanitization: There is no evidence of payload validation or sanitization in the provided logic for handling incoming events.
Audit Metadata