iblai-api-agent-sandbox
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill communicates exclusively with vendor-owned infrastructure at api.iblai.app to manage agent sandboxes. All network operations are part of the intended platform functionality.
- [COMMAND_EXECUTION]: The skill documents administrative commands for self-hosted sandbox management, such as service status checks and device pairing approval. These instructions are provided for manual user execution and are not invoked automatically by the skill's code.
- [CREDENTIALS_UNSAFE]: Secure management of sensitive data is implemented by treating gateway tokens and device identity keys as write-only fields, preventing their exposure through API read requests.
- [EXTERNAL_DOWNLOADS]: The documentation references official vendor repositories on GitHub for configuration examples and installation guides, all within the vendor's verified namespace.
- [INDIRECT_PROMPT_INJECTION]: While the skill manages agent prompt components like behavioral guidelines, it includes explicit security controls such as a configuration deny-list. This prevents the modification of sensitive system parameters, including sandbox enforcement and host-level execution settings.
Audit Metadata