Skip to main content
Integrations · Model providersDecides before data leaves

OpenAI-compatibleSupported calls, on the RedactWall route.

Compatible applications you explicitly configure to use RedactWall send supported OpenAI-style requests through the gateway. Requests and responses on that route are checked before forwarding. Direct provider traffic remains outside the gateway, as do unsupported protocol features.

In practice

What that looks like

A routed internal app sends a supported request containing a customer’s SSN

Stopped before it reaches the provider.

A supported response returns through RedactWall with a routing number

Caught before the application displays it to staff.

A team swaps one model vendor for another

The approved policy follows the workflow after the compatible application is configured to use the RedactWall route.

What it takes to switch on

Configure each compatible application to use the RedactWall gateway endpoint for supported OpenAI-style traffic. Only requests and responses that use that route are governed.

How far this is proven

Built to the vendor’s own connection and tested

RedactWall is built to OpenAI’s own connection and tested against it. This broader product integration is outside the proposed first pilot. The first run against a live provider account remains to be verified separately.

Every limit, in one place: Disclosures.

Review the proposed first pilot

The proposed first pilot covers ChatGPT text in managed Chrome and a separately verified Anthropic gateway path. Other integrations are part of the broader product.

  • Coverage is published for this integration
  • Proof status is shown explicitly
  • Pilot scope is agreed before onboarding