manage-lexicon

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements robust data governance practices. Destructive actions in commands/reset-lexicon.md require a literal, case-sensitive CONFIRM string, and all enrichment or reset tasks require a manual preview and confirmation step before execution.
  • [SAFE]: The skill maintains an audit trail by writing JSON logs to a local data-governance-runs/ directory after any write operation, ensuring accountability for metadata changes.
  • [SAFE]: The skill respects data privacy and system integrity by explicitly excluding Mixpanel-managed reserved namespaces (properties starting with mp_ or $) and legacy mobile SDK events from its working set.
  • [SAFE]: External communication is limited to the official Mixpanel MCP integration. The JSON query found in assets/volume-ranking-query.json is a standard Mixpanel Insights query for event volume ranking and contains no malicious payloads.
  • [SAFE]: No obfuscation, persistence mechanisms, or unauthorized privilege escalations were detected. The skill operates within the scoped context of a Mixpanel project as intended by the author.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 04:06 AM