aigw-auth

Configure BackendSecurityPolicy to attach provider credentials to AIServiceBackend or InferencePool.

3|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/missBerg/envoy-skills --skill aigw-auth
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aigw-auth
Source: https://github.com/missBerg/envoy-skills/tree/main/ai-gateway/adopters/skills/aigw-auth
Command: npx skills add https://github.com/missBerg/envoy-skills --skill aigw-auth

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Configure BackendSecurityPolicy for AI provider authentication to secure AIServiceBackend and InferencePool integrations across Envoy Gateway.

Core Features & Use Cases

  • Attach to AIServiceBackend or InferencePool to inject credentials for provider requests.
  • Support multiple credential types: APIKey, AnthropicAPIKey, AzureAPIKey, AWSCredentials, GCPCredentials.
  • Enforce one policy per target backend to avoid reconciliation conflicts.

Quick Start

Attach a BackendSecurityPolicy to your AIServiceBackend using a supported credential type (e.g., APIKey) and define the corresponding secret references.

Frequently Asked Questions about aigw-auth

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure BackendSecurityPolicy for AI provider authentication?

Configure BackendSecurityPolicy for AI provider authentication by defining targetRefs, selecting a credential type like APIKey or AWSCredentials, and providing the correct secret references for your chosen provider to secure access.

What credential types are supported for AI backend authentication in Envoy Gateway?

AI backend authentication in Envoy Gateway supports APIKey, AnthropicAPIKey, AzureAPIKey, AWSCredentials, and GCPCredentials, allowing you to inject the appropriate provider-specific credentials into requests.

Can I attach multiple authentication policies to a single AIServiceBackend?

No, you cannot attach multiple authentication policies to a single AIServiceBackend because a one-policy-per-target constraint is enforced to avoid reconciliation conflicts across your AI backends.

How do I secure an InferencePool using AWSCredentials?

Secure an InferencePool using AWSCredentials by attaching a BackendSecurityPolicy that specifies AWSCredentials as the credential type and references the correct AWS secret for provider requests.

Why does my BackendSecurityPolicy fail to apply across multiple AI backends?

Your BackendSecurityPolicy may fail if it violates the one-policy-per-target constraint, which prevents reconciliation conflicts when applying authentication configurations across multiple AI backends.

Does BackendSecurityPolicy work with both AIServiceBackend and InferencePool?

Yes, BackendSecurityPolicy works with both AIServiceBackend and InferencePool, allowing you to inject credentials and enforce secure AI provider access across different backend integration types.