byok-custom-model
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill implements model management logic by executing a Python script within a shell environment to update the local model registry and interact with system services.
- [EXTERNAL_DOWNLOADS]: The underlying script performs network operations to the local host (loopback) to synchronize the agent's state and may make external requests to verified LLM provider endpoints to fetch live model catalogs and pricing details.
- [SAFE]: The skill enforces a secure multi-step onboarding process for API keys, utilizing a specialized environment variable tool to ensure credentials are never processed in plain text or stored within conversation logs.
Audit Metadata