skills.rest
neondatabase
neondatabaseOFFICIAL
Explore all skills by this author →
Website𝕏@neondatabaseGitHub

neon-ai-gateway

Route LLM inference requests through a unified Neon database branch endpoint.

78|38|Updated Oct 20, 2022
Search Tags:#agent development#ai integration#neon#model routing#openai sdk#serverless ai#llm gateway
One-click install
npx skills add https://github.com/neondatabase/examples --skill neon-ai-gateway
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: neon-ai-gateway
Source: https://github.com/neondatabase/examples/tree/main/with-ai-sdk/.agents/skills/neon-ai-gateway
Command: npx skills add https://github.com/neondatabase/examples --skill neon-ai-gateway

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Juggling separate API keys, accounts, and code rewires to switch between LLM providers like Claude, GPT, and Gemini is tedious, and AI requests often lack the same branch-level isolation as your Neon database data.

Core Features & Use Cases

  • Unified LLM endpoint: Access frontier and open-source models from 7+ providers (Anthropic, OpenAI, Google, Meta, Alibaba, DeepSeek, Databricks) with a single Neon credential, no separate provider accounts or keys required.
  • Cross-provider model switching: Change only the model ID in your existing SDK code to move between providers, no rewiring needed for OpenAI, Anthropic, Google GenAI, Vercel AI SDK, or Mastra.
  • Branch-scoped AI isolation: Each Neon branch has its own dedicated gateway endpoint, so preview, CI, and feature branch AI requests are fully isolated from production, matching your database branch workflow.
  • Use Case: A development team building a customer support chatbot can test different LLM providers on preview branches without managing separate API keys, and keep test AI requests fully isolated from production traffic.

Quick Start

Use the neon-ai-gateway skill to enable the AI Gateway on your Neon branch and send a test chat completion request to the GPT-5-mini model using your existing OpenAI SDK setup.

Frequently Asked Questions about neon-ai-gateway

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

FAQPage Schema
How do I use multiple LLM providers without managing separate API keys?▼

A unified LLM endpoint allows you to access models from multiple providers using a single credential. This eliminates the overhead of managing separate API keys and accounts for different LLM providers in your application.

Can I use the OpenAI SDK to call models from Anthropic or Google?▼

Yes, you can use the OpenAI SDK with other providers by modifying only the base URL. This allows you to switch between different frontier and open-source models by changing only the model ID in your existing code.

Does the Neon AI gateway support branch-scoped request isolation?▼

Yes, the gateway supports branch-scoped AI request isolation. Each Neon database branch has its own dedicated gateway endpoint, ensuring that preview, CI, and feature branch AI requests are fully isolated from production traffic.

What's the best way to switch between LLM models in a serverless application?▼

Cross-provider model switching is the best way to switch models in serverless applications. You can move between providers like Anthropic, OpenAI, and Google by changing only the model ID, without needing to rewrite your code.

Do I need additional infrastructure for streaming long-form agent responses?▼

No, you do not need additional infrastructure for streaming long-form agent responses. The unified LLM endpoint provides native streaming support, allowing you to handle agent tool calling and long-form responses directly.

Related Skills

Explore More Software Engineering Skills→
legions-developerlegions-developer
885

neon-postgres

Guide Neon Serverless Postgres setup, connection methods, and CLI automation.

Community
Intermediate
botPhUbotPhU

ai-sdk-expert

Design AI applications with Vercel AI SDK v5 streaming and tool calling.

Community
Advanced
GGEduGGEdu

claude-api

Integrate Claude API applications using Python and TypeScript SDKs.

Community
Advanced
paarths-collabpaarths-collab
13

hermes-model

Configure remote Hermes model and provider settings over SSH with rollback.

Community
Advanced