prepare-ai-readiness

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows least-privilege principles by delegating data layer enrichment to the specialized manage-lexicon skill rather than requesting broad permissions to modify the Mixpanel schema directly.
  • [SAFE]: Implements strong user control mechanisms, including a mandatory 'Preview + diff' flow and an explicit CONFIRM string requirement before performing any write operations to the Mixpanel environment.
  • [SAFE]: Incorporates data integrity practices by requiring backups (to local files) of existing business context before it is replaced, ensuring a recovery path for users.
  • [SAFE]: The skill restricts its scope to documented Mixpanel business context APIs and requires proper role-based access control (Admin/Owner) as defined by the platform's security model.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 01:46 PM