kafka

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to install the @platformatic/kafka package from the public npm registry. This is a legitimate dependency provided by a well-known organization in the Node.js ecosystem.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Risk.
  • Ingestion points: Files references/platformatic-kafka.md and references/usage-guide.md provide instructions for the agent to consume data from external Kafka topics using the Consumer.consume() method.
  • Boundary markers: The instructions lack specific guidance on using delimiters or "ignore instructions" markers when the agent processes the content of consumed Kafka messages.
  • Capability inventory: The skill equips the agent with capabilities to manage Kafka infrastructure, such as Admin.createTopics and Producer.send (documented in references/platformatic-kafka.md), which could be triggered or abused if a malicious Kafka message influences the agent's behavior.
  • Sanitization: No mention of message payload validation, schema enforcement (beyond conceptual mentions of a registry), or escaping is present in the implementation instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 08:30 AM
Security Audit — agent-trust-hub — kafka