Agent Skills: Cloudflare Deployment Skill

Cloudflare Deployment Skill

When to Activate

Activate this skill when:

• Setting up Cloudflare Workers or Pages
• Working with KV, R2, or D1 storage
• Deploying applications to Cloudflare
• Configuring wrangler.toml
• Managing Cloudflare resources

Quick Commands

# Install wrangler
pnpm add -g wrangler

# Login
wrangler login

# Initialize new Worker
wrangler init my-worker

# Local development
wrangler dev

# Deploy to Cloudflare
wrangler deploy

# View logs
wrangler tail my-worker

Service Overview

| Service | Purpose | Use Case | | ----------- | ------------------------ | ------------------------- | | Workers | Serverless functions | API endpoints, middleware | | Pages | Static sites + functions | SvelteKit, Next.js | | KV | Key-value storage | Caching, session data | | R2 | Object storage | Files, images, backups | | D1 | SQLite database | Structured data |

Workers Setup

wrangler init my-worker --type javascript
cd my-worker
wrangler dev
wrangler deploy

Basic Worker

export default {
	async fetch(request, env) {
		const url = new URL(request.url);

		if (url.pathname === "/api/hello") {
			return new Response(JSON.stringify({ message: "Hello!" }), {
				headers: { "Content-Type": "application/json" },
			});
		}

		return new Response("Not found", { status: 404 });
	},
};

KV Storage

# Create namespace
wrangler kv namespace create MY_KV

# Add to wrangler.toml
# kv_namespaces = [{ binding = "MY_KV", id = "abc123" }]

// In Worker
export default {
	async fetch(request, env) {
		// Write
		await env.MY_KV.put("key", "value");

		// Read
		const value = await env.MY_KV.get("key");

		return new Response(value);
	},
};

R2 Storage

wrangler r2 bucket create my-bucket
# Add: r2_buckets = [{ binding = "MY_BUCKET", bucket_name = "my-bucket" }]

// In Worker
await env.MY_BUCKET.put("file.txt", "content");
const object = await env.MY_BUCKET.get("file.txt");
const text = await object.text();

D1 Database

wrangler d1 create my-database
wrangler d1 execute my-database --command="CREATE TABLE users (id INTEGER PRIMARY KEY, name TEXT)"
# Add: d1_databases = [{ binding = "DB", database_name = "my-database", database_id = "..." }]

// In Worker
const { results } = await env.DB.prepare("SELECT * FROM users WHERE id = ?").bind(1).all();

return Response.json(results);

Infra SDK Abstraction Layer

Grove uses portable infrastructure interfaces from @autumnsgrove/infra instead of raw Cloudflare APIs. This enables backend portability and consistent patterns across all services.

| Raw Cloudflare | Infra SDK Interface | Import | | ---------------------------- | ------------------- | --------------------- | | env.DB (D1Database) | GroveDatabase | @autumnsgrove/infra | | env.STORAGE (R2Bucket) | GroveStorage | @autumnsgrove/infra | | env.CACHE_KV (KVNamespace) | GroveKV | @autumnsgrove/infra | | Service bindings (Fetcher) | GroveServiceBus | @autumnsgrove/infra |

Creating a Grove Context

Create a GroveContext via createCloudflareContext() in your Worker entry point:

import { createCloudflareContext } from "@autumnsgrove/infra/cloudflare";

const ctx = createCloudflareContext({
	db: env.DB,
	storage: env.STORAGE,
	kv: env.CACHE_KV,
});

// Use portable interfaces
const files = await ctx.storage.list("prefix");
const cached = await ctx.kv.get("key");

Storage Management with Amber SDK

For file/storage management, use Amber SDK (@autumnsgrove/lattice/amber) which provides FileManager, QuotaManager, ExportManager, and AddonManager on top of GroveStorage:

import { FileManager, QuotaManager } from "@autumnsgrove/lattice/amber";

const fileManager = new FileManager(ctx.storage);
const quotaManager = new QuotaManager(ctx.storage);

// Type-safe file operations
const metadata = await fileManager.upload("user-123/photo.jpg", stream);
const quota = await quotaManager.getUsage("user-123");

Type-Safe Data Access

Use Rootwork utilities from @autumnsgrove/lattice/server for boundary type safety:

import { safeJsonParse, parseFormData } from "@autumnsgrove/lattice/server";
import { z } from "zod";

// Parse KV reads safely
const cached = await ctx.kv.get("config");
const config = safeJsonParse(cached, ConfigSchema);

// Parse form data with validation
const formData = await request.formData();
const data = parseFormData(formData, UploadSchema);

Pages Deployment

# Deploy static site
wrangler pages deploy ./build --project-name=my-site

# With SvelteKit
pnpm add -D @sveltejs/adapter-cloudflare
pnpm build
wrangler pages deploy .svelte-kit/cloudflare

wrangler.toml Configuration

name = "my-worker"
main = "src/index.js"
compatibility_date = "2024-01-01"

# Automatic resource provisioning (v4.45.0+)
kv_namespaces = [{ binding = "MY_KV" }]
r2_buckets = [{ binding = "MY_BUCKET" }]
d1_databases = [{ binding = "DB" }]

Development Modes

# Local (simulated resources)
wrangler dev

# Remote (real Cloudflare resources)
wrangler dev --remote

Worker Secrets

# Set secret
wrangler secret put API_KEY --name my-worker

# Access in code
const apiKey = env.API_KEY;

Best Practices

DO ✅

• Use wrangler dev for local testing first
• Use environment-specific configs
• Monitor logs with wrangler tail
• Use automatic resource provisioning

DON'T ❌

• Hardcode account IDs
• Skip local testing
• Commit wrangler.toml with production IDs
• Ignore rate limits

Common Patterns

KV + R2 Caching

export default {
	async fetch(request, env) {
		const url = new URL(request.url);
		const cacheKey = url.pathname;

		// Check KV cache
		let content = await env.MY_KV.get(cacheKey);
		if (content) return new Response(content);

		// Fetch from R2
		const object = await env.MY_BUCKET.get(cacheKey.slice(1));
		if (!object) return new Response("Not found", { status: 404 });

		content = await object.text();

		// Cache in KV
		await env.MY_KV.put(cacheKey, content, { expirationTtl: 3600 });

		return new Response(content);
	},
};

Related Resources

See AgentUsage/cloudflare_guide.md for complete documentation including:

• MCP server configuration
• Advanced D1 patterns
• Production deployment strategies
• Troubleshooting guide