Free IDE extension for risk-free vibe coding — keep secrets out of AI.
Control
POST /api/escrow/createEscrow service for AI agent actions that require human approval. Risky actions are held in escrow with full context, allowing human reviewers to approve, deny, or modify before execution.
Complete these steps in order. The dashboard link opens the exact product workspace for this service.
Sign in, select the project you want to protect, then open Transaction Escrow.
When an agent action exceeds the auto-approval threshold, it's intercepted and placed in escrow.
Use POST /api/escrow/create from server-side code. Keep API keys out of browsers and client bundles.
Run one safe case and one expected-risk case. Confirm the result in Transaction Escrow, then review Guard Logs before enabling the control for production traffic.
Some actions are too risky to allow automatically but too important to simply block. Transaction Escrow enables a human-in-the-loop workflow where subject matter experts review and approve sensitive actions before they're executed.
A step-by-step breakdown of how Transaction Escrow protects your AI.
When an agent action exceeds the auto-approval threshold, it's intercepted and placed in escrow.
Full context is preserved — including the action details, content, destination, and risk assessment.
Authorized reviewers receive notifications and can review the action through the dashboard or API.
Reviewers can approve, deny, edit content, or set time-based conditions for the action.
See the complete action context including content, destination, and risk scoring
Modify action content before approval to add safe versions
Set expiration times for approvals to prevent stale actions
Approve or deny multiple escrowed actions at once
This service is implemented at the application route below. Use the tested SDK and framework guides for copy-paste integration examples.
POST /api/escrow/create. API keys must stay server-side. See the REST API guide and related documentation before integrating.