skills/builderio/agent-native/dictate/Gen Agent Trust Hub

dictate

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes pnpm action commands for application navigation and performing text cleanup tasks. These actions are standard operational procedures within the Clips framework and are used as intended for feature integration.\n- [SAFE]: Managed credentials and user-provided API keys (BYOK) are handled through established configuration paths. The skill correctly prioritizes the managed 'Builder.io Connect' service and mentions environment-based keys only as a fallback.\n- [SAFE]: Access to sensitive dictation data is governed by internal security functions such as accessFilter and assertAccess, which mitigate the risk of unauthorized data exposure.\n- [SAFE]: The skill's instructions specifically prohibit the agent from blocking user interactions with network round-trips or deleting data without explicit confirmation, following user-centric safety best practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — dictate