twenty-engine-sync

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements identity resolution and data merge logic using pure Python within the provided demo files. It does not attempt to execute remote code or download untrusted dependencies.
  • [SAFE]: Security is managed through an 'operator-gated' workflow. All commands that modify CRM data are restricted to dry-runs by default and require explicit human approval before execution.
  • [SAFE]: No instances of prompt injection, data exfiltration, or hardcoded credentials were found in the instructions or the accompanying code.
  • [SAFE]: The skill follows least-privilege principles by focusing on additive updates and prohibiting destructive actions like record deletion or schema modification.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 05:02 PM
Security Audit — agent-trust-hub — twenty-engine-sync