nocobase-ai-manager

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the nb CLI for platform management, incorporating mandatory secondary confirmation steps for all disruptive actions such as service deletion or disabling.
  • [PROMPT_INJECTION]: To protect against indirect injection from ingested platform data, the skill enforces strict manual verification gates and independent readback of configuration states before proceeding with dependent tasks.
  • [REMOTE_CODE_EXECUTION]: Dynamic generation of JSON data payloads is handled securely using temporary files with restricted owner-only permissions (umask 077) and automated cleanup using shell traps.
  • [DATA_EXFILTRATION]: The skill follows industry best practices for data protection by forbidding the exposure of secrets in process listings, history, or command summaries, and by masking sensitive provider options in all outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 07:30 AM
Security Audit — agent-trust-hub — nocobase-ai-manager