agents-sdk

Develop stateful AI agents on Cloudflare Workers using the Agents SDK.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/JuanJoseGonGi/skills --skill agents-sdk-juanjosegongi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agents-sdk
Source: https://github.com/JuanJoseGonGi/skills/tree/main/agents-sdk
Command: npx skills add https://github.com/JuanJoseGonGi/skills --skill agents-sdk-juanjosegongi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a robust framework for building stateful, persistent AI agents that run on Cloudflare's edge, enabling complex applications like real-time chat, scheduled tasks, and durable workflows.

Core Features & Use Cases

  • Stateful Agents: Maintain context and data across requests and reconnections using Durable Objects.
  • Real-time Communication: Supports WebSockets and HTTP for seamless client-agent interaction.
  • Durable Workflows & Scheduling: Orchestrate multi-step tasks and schedule recurring jobs.
  • Tool Calling & RPC: Integrate AI models with external tools and expose agent methods as callable APIs.
  • Use Case: Develop a customer support chatbot that remembers past conversations, can schedule follow-up actions, and integrates with your CRM via tool calls.

Quick Start

Use the agents-sdk skill to build a new AI agent by running npm create cloudflare@latest -- my-agent --template=cloudflare/agents-starter.

Frequently Asked Questions about agents-sdk

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build stateful AI agents on Cloudflare Workers?

Stateful AI agents on Cloudflare Workers are built using the Agents SDK, which leverages Durable Objects to maintain context and data across requests. It supports real-time WebSockets, scheduled tasks, and tool calling for complex edge applications.

How does Cloudflare Durable Objects handle state management for AI agents?

Durable Objects handle state management by persisting agent context and data across reconnections. This allows AI agents to maintain memory of past interactions and execute recurring scheduled jobs without losing state.

Can I use WebSockets and RPC for real-time communication with AI agents?

Yes, WebSockets and RPC facilitate real-time communication with AI agents. The framework supports seamless client-agent interaction over WebSockets and exposes agent methods as callable RPC APIs for external integration.

Does the Agents SDK support scheduling recurring tasks on Cloudflare?

The Agents SDK supports scheduling recurring tasks on Cloudflare by orchestrating durable workflows and multi-step scheduled jobs. This enables agents to execute automated follow-up actions and integrate with Cloudflare Workflows.

What is the best way to integrate external tools with AI models on Cloudflare edge?

Integrating external tools with AI models on Cloudflare edge is best achieved using the Agents SDK's tool calling feature. It connects AI models with external systems like CRMs and exposes agent methods as callable APIs.

Why do I need Cloudflare Workers for persistent AI chat applications?

Cloudflare Workers are needed for persistent AI chat applications because they provide an edge environment supporting stateful agents. Durable Objects ensure chatbots remember past conversations and handle WebSocket reconnections seamlessly.