Introduction
What AI Gateway is and the risks it addresses.
AI Gateway is an AI security gateway for OpenAI-compatible and Anthropic APIs. It scans outbound prompts and inbound responses against the Cedar policy managed by Policy. It is installed as a single umbrella Helm chart that bundles the gateway and its runtime dependencies.
Start with the Introduction for an overview, then read the Architecture to understand the gateway and its deployment components.
Follow the Installation guide, then use Uninstallation to manage the deployment lifecycle. The API Reference documents gateway endpoints, Configuration provides policy and environment-variable examples, Usage covers identity and client connections, and Troubleshoot helps diagnose common issues.
What AI Gateway is and the risks it addresses.
Overview of AI Gateway architecture, processing flow, and deployment components on AWS EKS.
Prepare and install AI Gateway on a supported Kubernetes cluster.
Example YAML policies and runtime environment variables for AI Gateway.
Remove the AI Gateway Helm release, clean up Kubernetes resources, and destroy cloud infrastructure with OpenTofu.
AI Gateway endpoints, input fields, and output fields.
Create users, connect applications, and operate AI Gateway.
Troubleshoot common AI Gateway deployment, identity, and scanner issues.
Was this page helpful?