Sandbox for AI-Generated Code

Review sandbox controls for AI-generated code execution workflows before production. This page uses the same rule-based risk checker and checklist path.

TOOL_RC_01

Risk Assessment Entry

Describe capabilities, not sensitive implementation details. The rules engine will return your review checklist.

Agent typeExecution capability

Access scope

Current isolationDeployment context

No secrets. Do not submit API keys, credentials, source code, private logs, or customer data.

Your agent risk profile

Your risk checklist will appear here after you describe your agent and run the check.