hive-build-onboarding

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to execute npx -y -p hive-intelligence@latest hive init --browser. This fetches the official CLI tool from the vendor's package repository on npm. This is a standard and safe practice for onboarding CLI tools from a trusted vendor.
  • [COMMAND_EXECUTION]: Provides commands for the user to run manually (hive init, hive doctor, curl). These commands are used for local configuration and verification of the vendor's own services. No autonomous or hidden command execution was detected.
  • [DATA_EXFILTRATION]: The skill explicitly instructs users not to paste secrets into chat or logs. It describes a secure authentication flow where the API key is passed directly from a local browser redirect to a local CLI listener, ensuring the secret remains outside the agent's context and logs.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The documentation correctly handles placeholders like hive_live_… and HIVE_API_KEY, and it provides accurate advice on revoking lost keys and using secret managers.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 12:27 AM
Security Audit — agent-trust-hub — hive-build-onboarding