kibana

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). Kibana actions are executed via ClawLink, which at runtime reads Kibana data (e.g., alert/case/saved-object contents) and feeds the retrieved text back into the agent’s LLM context; that content is authored/maintained by Kibana users other than the operating user (outsider source: authenticated app data/records).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill invokes the ClawLink CLI via npx (package page: https://www.npmjs.com/package/@useclawlink/cli and service https://claw-link.dev), which causes remote code to be fetched and executed at runtime and is required for the skill to operate.