Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.edgespark.dev/llms.txt

Use this file to discover all available pages before exploring further.

EdgeSpark is a platform for deploying full-stack web applications to Cloudflare’s edge network. You write a Hono server in TypeScript, define your app schema in the repo, and use EdgeSpark for deployment, authentication, storage, secrets, and runtime wiring. EdgeSpark is designed to be agent-native: AI agents can scaffold, build, and deploy complete applications with minimal friction. If you want to evaluate EdgeSpark quickly, start with the quickstart, verify the CLI installation, explore how EdgeSpark works, and check the supported AI agents.

​
What you get

Repo-owned database schema

Define tables with Drizzle ORM in server/src/defs/, then generate and apply migrations with the CLI.

Authentication

Built-in user auth with path-based enforcement and configurable auth settings.

R2 storage

Declare buckets in code, apply them with the CLI, and use typed storage helpers at runtime.

Secrets and vars

Manage encrypted secrets and plain env vars through the CLI and read them from the runtime SDK.

​
How it works

You write a standard Hono app and import the runtime SDK where you need it:
server/src/index.ts
import { db } from "edgespark";
import { auth } from "edgespark/http";
import { Hono } from "hono";
import { posts } from "@defs";

const app = new Hono().get("/api/posts", async (c) => {
  const rows = await db.select().from(posts);
  return c.json({ posts: rows, userEmail: auth.user.email });
});

export default app;
No database credentials, storage clients, or auth middleware setup is required in your app code.

​
Built for AI agents

EdgeSpark is a harness platform. It gives agents a predictable scaffold, verified CLI workflows, and generated types that keep code aligned with the deployed project. AI agents are first-class users of EdgeSpark. EdgeSpark supports all AI agents: Claude Code, Gemini CLI, OpenAI Codex, GitHub Copilot, and Cursor through dedicated plugins or extensions, plus OpenCode, Amp, Devin, Aider, Windsurf, Cline, Continue, Antigravity, Kiro, and every other AI agent through the generic EdgeSpark agent-skills plus docs MCP path. The For AI Agents tab is the agent operating manual.

​
Why teams choose EdgeSpark

EdgeSpark is for teams that want the speed of AI-assisted development without giving up control over their production architecture. Your app schema, storage buckets, auth config, and runtime keys stay in the repo, while the platform handles deployment, generated types, and the guardrails around auth, SQL, and storage access. This makes EdgeSpark different from a framework-only setup. You still write a normal Hono and TypeScript application, but the platform gives you a safer path to ship full-stack apps on Cloudflare Workers, D1, and R2 with less custom plumbing.

​
Common EdgeSpark use cases

  • Deploy a Hono API to Cloudflare’s edge with D1, authentication, R2 storage, secrets, and typed runtime imports
  • Give Claude Code, Gemini CLI, OpenAI Codex, GitHub Copilot, Cursor, OpenCode, Amp, Devin, Aider, Windsurf, Cline, Continue, Antigravity, Kiro, and all other AI agents a predictable full-stack scaffold
  • Ship production changes with dry runs, generated types, schema migrations, and a deploy pipeline that matches the live platform

​
Get started

Quickstart

Deploy your first project in minutes.

How EdgeSpark works

Understand the platform and runtime model.

CLI reference

Verified public CLI commands and workflows.
Last modified on April 9, 2026