Deploying Hono on Cloudflare Workers

This is the last post in my Bun and Hono mini-series. We started with Bun, looked at Hono itself, then middleware. Now we deploy.

If you’ve been following my Cloudflare Workers series, you already know the platform. A Worker is a function that takes a request and returns a response. Hono is built for exactly that model.

Why Hono on Workers

Hono was born on Cloudflare Workers. The name means “flame” in Japanese, and the project started as a tiny framework for the edge.

Workers give you a fetch handler. Hono gives you routing, middleware, and helpers on top of that same handler. No adapter, no glue code.

The whole library is small and fast. That’s why it’s my go-to when I need a small API on Cloudflare.

Create the project

The Hono team ships a Cloudflare Workers template. Run the scaffolder:

npm create hono@latest my-api

When it asks for a template, pick cloudflare-workers. Or skip the prompt:

npm create hono@latest my-api -- --template cloudflare-workers

Move into the folder and install dependencies:

cd my-api
npm install

You get a minimal project. The entry point is src/index.ts:

import { Hono } from 'hono'

const app = new Hono()

app.get('/', (c) => {
  return c.text('Hello Hono!')
})

export default app

Hono exports the app directly. Cloudflare calls app.fetch under the hood.

The other important file is wrangler.jsonc:

{
  "$schema": "node_modules/wrangler/config-schema.json",
  "name": "my-api",
  "main": "src/index.ts",
  "compatibility_date": "2025-08-03"
}

That’s your Worker config. Bindings for D1, KV, or R2 go here too.

Run locally and deploy

The template wires up npm scripts for you. Start the local dev server:

npm run dev

That runs wrangler dev. Open http://localhost:8787 and you’ll see your hello message. Edit the code, save, and it reloads.

When you’re ready to ship:

npm run deploy

Wrangler bundles your code and uploads it. You get a *.workers.dev URL in seconds. The first time, it opens a browser so you can log in to Cloudflare.

Access bindings with types

On Workers, databases and storage show up on env. In Hono you read them from c.env.

Say you already created a D1 database and added it to wrangler.jsonc, like in my Cloudflare D1 post:

{
  "d1_databases": [
    {
      "binding": "DB",
      "database_name": "my-app-db",
      "database_id": "your-database-id"
    }
  ]
}

Pass the binding types to Hono as a generic:

type Bindings = {
  DB: D1Database
}

const app = new Hono<{ Bindings: Bindings }>()

Now c.env.DB is typed. You can also run npm run cf-typegen to generate types from your config automatically.

A tiny posts API

Let’s wire up two routes against a posts table. I won’t re-explain D1 here — check the D1 post for migrations and schema setup.

Read all posts:

app.get('/api/posts', async (c) => {
  const { results } = await c.env.DB.prepare(
    'select * from posts order by created_at desc'
  ).all()

  return c.json(results)
})

Create a post:

app.post('/api/posts', async (c) => {
  const { title, body } = await c.req.json()

  await c.env.DB.prepare(
    'insert into posts (title, body, created_at) values (?, ?, ?)'
  ).bind(title, body, Date.now()).run()

  return c.json({ ok: true }, 201)
})

That’s a real API. Routing, JSON parsing, and a database query — all in one file.

Middleware and other bindings

Middleware works the same on Workers as anywhere else. Cookies, headers, CORS — everything from the middleware post applies here unchanged.

Other bindings work the same way. Add a KV namespace to wrangler.jsonc, type it on Bindings, and use c.env.SESSIONS (or whatever you named it). Same pattern for R2 buckets. I covered KV in a separate post.

Wrapping up the series

We started with Bun as a fast runtime, learned Hono as a tiny web framework, explored middleware, and now we’ve deployed to the edge.

This is my preferred stack for small APIs on Cloudflare. Hono on Workers, D1 when I need SQL, KV when I need simple key lookups. Fast to scaffold, fast to deploy, fast at runtime.

If you want to go deeper, the Hono Cloudflare Workers guide and the Wrangler docs are the next stop.

Related posts about cloudflare: