zeroboot-vm-sandbox
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides isolated code execution environments using KVM-based VM technology. This architecture is specifically designed to mitigate the risks of executing untrusted code by using hardware-enforced boundaries.
- [EXTERNAL_DOWNLOADS]: The skill provides installation instructions for the official Zeroboot SDKs (zeroboot and @zeroboot/sdk) via standard package managers.
- [PROMPT_INJECTION]: Potential surface area for indirect prompt injection was evaluated because the skill executes agent-generated code. Ingestion points include the sb.run() method and API endpoints. The risk is mitigated by the sandbox's hardware isolation, which ensures guest code cannot escape to the host system. Sanitization is managed through physical environment separation rather than text filtering.
Audit Metadata