arkcli-custommodel

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements safety-first instructions, requiring the agent to verify user intent and check for active endpoint dependencies before executing destructive commands such as model deletion.
  • [COMMAND_EXECUTION]: The skill executes subcommands of the arkcli tool, which is the official command-line interface for the platform. These operations are restricted to managing custom models within the user's account and align with the skill's stated purpose.
  • [EXTERNAL_DOWNLOADS]: The skill references Volcengine's TOS (Object Storage) service for importing model weights. These references are to the vendor's own infrastructure and are used as data sources for model management, presenting no unusual security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 02:03 AM
Security Audit — agent-trust-hub — arkcli-custommodel