salesforce

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

• Third-party content exposure detected (medium risk: 0.30). The skill interacts with Salesforce CRM data and records via the Membrane CLI, where searching or retrieving records (such as leads, cases, or contacts) could ingest outsider-submitted content into the agent's context.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the installation and execution of an arbitrary third-party CLI package via npm from an uncontrolled source (@membranehq/cli), establishing a medium-tier operational dependency.