experience-cms-content-generate

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a 'No-drift STOP rule' which prevents the agent from attempting to use non-specialized tools (like SOQL or CLI) if the primary CMS MCP tools are unavailable, reducing the risk of accidental misconfiguration or privilege bypass.
  • [SAFE]: Content generation and editing are performed entirely on the server via the prompt parameter in the create_cms_content and update_cms_content_variant tools. This ensures that the agent does not handle the underlying content schema or perform client-side assembly of content, relying instead on server-side validation.
  • [SAFE]: The skill uses strict identifier validation based on recognized Salesforce ID prefixes (0Zu, 9Pu, 9Ps, MC, 20Y) and requires server confirmation via read tools before any write operation is attempted on a record identified by the user.
  • [SAFE]: Delegation to the experience-cms-content-type-generate skill is handled via a secure skill bridge and follows a well-defined protocol that pauses the parent skill and consumes only structured status responses, preventing state corruption.
  • [SAFE]: File operations are restricted to scanning local project directories for UIBundle metadata and persisting configuration to content-metadata.json within the local project structure, which is consistent with standard local development workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 11:21 PM
Security Audit — agent-trust-hub — experience-cms-content-generate