Cloudflare Agents

Name: Cloudflare Agents
Author: Claude Skills Maintainers

secondsky/claude-skills

196 repo stars
Updated July 25, 2026
secondsky/claude-skills

Build AI agents on Cloudflare Workers with MCP integration, tool use, and multiple LLM providers.

About

This skill covers building AI agents on Cloudflare Workers with MCP integration, tool use, and pluggable LLM providers. A developer uses it to ship agentic features that run at the edge and call tools, served through the Workers platform.

AI agents on Workers
MCP integration
Tool use & LLM providers
Edge-deployed via wrangler

Cloudflare Agents by the numbers

Data as of Jul 28, 2026 (Skillselion catalog sync)

/plugin marketplace add secondsky/claude-skills

/plugin install cloudflare-agents@claude-skills

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

[![Listed on Skillselion](https://skillselion.com/badge/skills/secondsky/claude-skills/plugins/cloudflare-agents.svg)](https://skillselion.com/skills/secondsky/claude-skills/plugins/cloudflare-agents)

repo stars	★ 196
Last updated	July 25, 2026
Repository	secondsky/claude-skills ↗

What it does

Build AI agents on Cloudflare Workers with MCP integration, tool use, and multiple LLM providers.

README.md

Cloudflare Agents SDK Skill

Status: Production Ready ✅ Last Updated: 2025-10-21 Production Tested: Cloudflare's own MCP servers (https://github.com/cloudflare/mcp-server-cloudflare)

Auto-Trigger Keywords

Primary Keywords (Core Technology):

cloudflare agents
agents sdk
cloudflare agents sdk
agents package
npm create cloudflare agents
npm i agents
build agents cloudflare
ai agents workers
durable objects agents
stateful agents

Secondary Keywords (API Surfaces):

Agent class, extend agent, onRequest agent, onConnect agent, onMessage agent
useAgent hook, AgentClient, agentFetch, useAgentChat
this.setState agent, agent state management, state sync agents, this.sql agent
this.schedule, schedule tasks agent, cron agents, agent alarms
run workflows agent, agent workflows, workflow integration agents
browse web agents, puppeteer agents, browser rendering agents
rag agents, vectorize agents, agent embeddings, retrieval augmented generation agents
mcp server, McpAgent, mcp tools, model context protocol
routeAgentRequest, getAgentByName, agent instance, calling agents
AIChatAgent, chat agents, streaming chat agent
websocket agents, server-sent events agents, sse agents

Use Case Keywords:

stateful micro-servers
long-running agents
autonomous agents
human in the loop agents, hitl agents, approval workflow agents
multi-agent systems, agent orchestration, agent communication
persistent agent state
agent with database
scheduled agent tasks
agent web scraping

Error-Based Keywords (Common Issues):

"Agent class must extend"
"new_sqlite_classes"
"migrations required"
"binding not found"
"agent not exported"
"callback does not exist"
"state limit exceeded"
"migration tag"
"workflow binding missing"
"browser binding required"
"vectorize index not found"
"Cannot enable SQLite on existing class"
"Migrations are atomic"

What This Skill Does

This skill provides comprehensive knowledge for the Cloudflare Agents SDK - a framework for building AI-powered autonomous agents on Cloudflare Workers + Durable Objects.

Agents can:

✅ Communicate in real-time via WebSockets and Server-Sent Events
✅ Persist state with built-in SQLite database (up to 1GB per agent)
✅ Schedule tasks using delays, specific dates, or cron expressions
✅ Run asynchronous Cloudflare Workflows
✅ Browse the web using Browser Rendering API + Puppeteer
✅ Implement RAG with Vectorize + Workers AI embeddings
✅ Build MCP (Model Context Protocol) servers
✅ Support human-in-the-loop patterns
✅ Scale to millions of independent agent instances globally

Each agent instance is a globally unique, stateful micro-server that can run for seconds, minutes, or hours.

Known Issues Prevented

This skill prevents 15+ documented issues:

Issue	Source	Impact
Migrations not atomic	Durable Objects Migrations	Cannot gradually deploy migrations - deployment fails
Missing new_sqlite_classes	Agents Configuration	Cannot enable SQLite later - must be in v1 migration
Agent class not exported	Agents API	Binding fails with "Cannot access undefined"
Binding name mismatch	Configuration	"Binding not found" error
Global uniqueness not understood	Agents API	Same name = same instance globally (can cause data leakage)
WebSocket state not persisted	WebSockets	Connection state lost after disconnect
Scheduled task callback missing	Schedule Tasks	Runtime error: "Method does not exist"
State size limit exceeded	State Management	Database exceeds 1GB limit
Task payload too large	Schedule Tasks	Task exceeds 2MB limit
Workflow binding missing	Run Workflows	Cannot access workflow binding
Browser binding required	Browse the Web	Puppeteer fails without binding
Vectorize index not found	RAG	Query fails on non-existent index
MCP transport confusion	MCP Transport	Using deprecated SSE instead of /mcp
Authentication bypass	Calling Agents	Security vulnerability
Instance naming errors	Calling Agents	Cross-user data leakage

When to Use This Skill

Use this skill when:

✅ Building AI-powered agents that need to persist state
✅ Creating chat applications with streaming responses
✅ Implementing real-time WebSocket communication
✅ Scheduling recurring tasks or delayed execution
✅ Running asynchronous workflows
✅ Building RAG systems with Vectorize + Workers AI
✅ Creating MCP servers for LLM tool use
✅ Implementing human-in-the-loop approval workflows
✅ Web scraping with headless browsers
✅ Building multi-agent systems
✅ Needing stateful micro-servers that scale globally
✅ Encountering Agent configuration or deployment errors

When NOT to Use This Skill

Don't use this skill when:

❌ Building simple stateless Workers (use cloudflare-worker-base instead)
❌ Only need key-value storage (use cloudflare-kv instead)
❌ Only need SQL database (use cloudflare-d1 instead)
❌ Building static websites (use cloudflare-worker-base with Static Assets)
❌ Project doesn't require persistent state or WebSockets

Quick Example

// Create an Agent class
import { Agent } from "agents";

interface Env {
  AI: Ai;
}

export class MyAgent extends Agent<Env> {
  async onRequest(request: Request): Promise<Response> {
    return Response.json({ agent: this.name, state: this.state });
  }

  async onConnect(connection: Connection, ctx: ConnectionContext) {
    connection.send("Welcome!");
  }

  async onMessage(connection: Connection, message: WSMessage) {
    if (typeof message === 'string') {
      connection.send(`Echo: ${message}`);
    }
  }
}

export default MyAgent;

// wrangler.jsonc
{
  "durable_objects": {
    "bindings": [{"name": "MyAgent", "class_name": "MyAgent"}]
  },
  "migrations": [
    {"tag": "v1", "new_sqlite_classes": ["MyAgent"]}
  ]
}