skills.rest
Moisesjr20
Moisesjr20COMMUNITY
Explore all skills by this author →
GitHub

ai-gateway

Route AI requests through Vercel AI Gateway with failover and per-user rate limits.

1|Updated Mar 10, 2026
Search Tags:#vercel#failover#rate-limiting#oidc#ai-gateway#model-routing#cost-tracking
One-click install
npx skills add https://github.com/Moisesjr20/gueclaw --skill ai-gateway-moisesjr20
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-gateway
Source: https://github.com/Moisesjr20/gueclaw/tree/main/.agents/skills/ai-gateway
Command: npx skills add https://github.com/Moisesjr20/gueclaw --skill ai-gateway-moisesjr20

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the complexity of managing multiple AI providers by providing a single, unified gateway for routing requests, handling provider failover, and attributing costs so teams can operate AI features reliably in production.

Core Features & Use Cases

  • Unified Model Routing: Use provider/model strings to route requests through Vercel AI Gateway for consistent provider selection.
  • Failover & Reliability: Configure provider order and fallback models to maintain uptime when a provider experiences errors or quotas.
  • Cost Controls & Observability: Per-user tags, budget alerts, and usage logs enable cost tracking, budget enforcement, and audit trails for multi-tenant applications.
  • Auth & Local Development: OIDC-first authentication via vercel env pull with AI_GATEWAY_API_KEY fallback supports both Vercel deployments and local CI usage.
  • Use Case: Deploy a multi-tenant chat service that requires per-user rate limits, per-feature cost attribution, and automatic failover between OpenAI and Anthropic.

Quick Start

Route the model openai/gpt-5.4 through the Vercel AI Gateway with per-user tags and a failover order of openai then anthropic.

Frequently Asked Questions about ai-gateway

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

FAQPage Schema
How do I set up unified model routing and provider failover for AI requests?▼

Unified model routing directs AI requests through a single gateway using provider/model strings, while failover configures a provider order to automatically fallback to alternative models during errors or quota limits.

What is the best way to enforce per-user rate limiting and cost tracking in a multi-tenant SaaS?▼

Per-user rate limiting and cost tracking in a multi-tenant SaaS are enforced by applying per-request tags, monitoring usage logs, and configuring programmatic budget alerts to maintain cost attribution.

How does OIDC authentication work with the Vercel AI Gateway for local development?▼

OIDC authentication operates via vercel env pull for secure deployments, with an AI_GATEWAY_API_KEY fallback mechanism specifically designed to support local CI usage and development workflows.

Can I route requests to specific models like OpenAI and Anthropic while maintaining uptime?▼

Routing requests to specific models like OpenAI and Anthropic maintains uptime by defining a failover order, ensuring the gateway automatically switches providers when the primary one experiences failures.

When do I need an AI gateway for my production application?▼

An AI gateway is needed for production applications requiring reliable provider selection, consistent cost attribution, audit logging, and automatic failover to manage multiple AI providers seamlessly.

Related Skills

Explore More Software Engineering Skills→
VollliebVolllieb

env-vars

Manages Vercel environment variables and OIDC tokens via CLI commands and .env files.

Community
Advanced
winsorllcwinsorllc

cost-tracker

Track LLM API costs across Anthropic, OpenAI, and Google providers.

Community
Intermediate
MohammedHTahirMohammedHTahir

ai-gateway

Route AI model calls through Vercel AI Gateway with failover and rate limiting.

Community
Advanced
vercelvercel
246

ai-gateway

Route AI model calls through a unified gateway with OIDC authentication.

Official
Advanced