Deploy Serverless Edge Functions with Cloudflare Workers

This skill provides senior-level Cloudflare Workers engineering guidance for edge computing architectures, performance optimization at the edge, and the full Cloudflare developer ecosystem (Wrangler, KV, D1, Queues, Durable Objects).

It covers the Wrangler ecosystem (using wrangler.toml for configuration and npx wrangler dev for local testing), the Fetch API (Workers use the Web standard Fetch API, not Node.js globals), bindings (defining KV, D1, and secrets in wrangler.toml and accessing them via the env parameter in the fetch handler), cold starts (Workers have 0ms cold starts but bundle size must stay within the free tier's 1MB limit), Durable Objects (for stateful coordination and high-concurrency needs), and error handling (waitUntil() for non-blocking async tasks like logging or analytics that run after the response is sent).

Best practices: use env.VAR_NAME for secrets and environment variables, use Response.redirect() for edge-side redirects, use wrangler tail for live production debugging; avoid importing large libraries (limited memory/CPU time) and avoid Node.js-specific libraries like fs or path unless using Node.js compatibility mode. It documents a specific troubleshooting case: a request exceeding the CPU time limit is fixed by optimizing loops, reducing await calls, moving heavy synchronous work out of the request/response path, and using ctx.waitUntil() for non-blocking tasks.

export interface Env {
  MY_KV_NAMESPACE: KVNamespace;
}

export default {
  async fetch(
    request: Request,
    env: Env,
    ctx: ExecutionContext,
  ): Promise<Response> {
    const value = await env.MY_KV_NAMESPACE.get("my-key");
    if (!value) {
      return new Response("Not Found", { status: 404 });
    }
    return new Response(`Stored Value: ${value}`);
  },
};

Use this skill when designing and deploying serverless functions to Cloudflare's edge, implementing edge-side data storage with KV/D1/Durable Objects, optimizing latency by moving logic to the edge, building full-stack apps with Cloudflare Pages and Workers, or handling request/response modification, security headers, and edge-side caching.

Not for traditional Node.js/Express apps run on servers, AWS Lambda or Google Cloud Functions (use their respective skills), or general frontend development that doesn't use edge features.

Inputs: a serverless/edge function requirement - request handling, data storage, response modification, or full-stack app architecture on Cloudflare.

Outputs: a Worker (with wrangler.toml configuration and bindings) implementing the required edge logic, using KV/D1/Durable Objects as needed, following Fetch API and CPU/bundle-size constraints.

Wrangler CLI, Cloudflare KV, D1, Durable Objects, Queues, Cloudflare Pages.

Developers building serverless functions, edge-side data storage, or full-stack applications on Cloudflare Workers and Pages.