What problem does it solve? Lambda-style serverless functions time out on long-running workloads like AI agents, WebSocket servers, and SSE streams, and they run far from your database. This Skill guides you through deploying long-running Node.js functions directly onto a Neon branch, with DATABASE_URL injected automatically and compute running in the same region as your Postgres data. ## Core Features & Use Cases - Long-Running Handlers: Deploy web-standard fetch handlers with a 15-minute time-to-first-byte budget, supporting streaming AI agents, WebSocket servers, and SSE endpoints that outlast traditional serverless limits. - Branch-Scoped Backends: Each Neon branch runs its own function version at its own URL against its own isolated database, so preview deployments and CI get self-contained backends. - Function Triggers: Configure cron schedules or object-storage events that POST to your function for recurring HTTP work without a separate scheduler. - Use Case: Build an AI agent with the Vercel AI SDK that streams multi-step tool-calling responses, queries Postgres through a module-scope connection pool, and authenticates callers with JWT verification — all deployed with neon deploy. ## Quick Start Ask the AI to declare a function in neon.ts with a Hono handler that queries Postgres via the injected DATABASE_URL, then run neon dev locally and neon deploy to get a public invocation URL.
