chatgpt-apps

Scaffold ChatGPT Apps SDK projects with MCP server and widget UI.

779|80|Updated Jun 20, 2025
One-click install
npx skills add https://github.com/Haohao-end/openagent --skill chatgpt-apps-haohao-end
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chatgpt-apps
Source: https://github.com/Haohao-end/openagent/tree/main/api/internal/core/skills/catalog/chatgpt-apps
Command: npx skills add https://github.com/Haohao-end/openagent --skill chatgpt-apps-haohao-end

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node:http, node:fs, node:path, node:url, zod, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Scaffolding ChatGPT Apps SDK projects often becomes a moving-target task because tool wiring, MCP Apps bridge patterns, and UI/resource metadata need to match current documentation and conventions.

Core Features & Use Cases

  • Docs-first ChatGPT Apps scaffolding: Plans tools and repo shape after pulling the latest Apps SDK build and quickstart guidance.
  • MCP server + widget generation: Produces an MCP server scaffold that registers widget resources/templates and tool handlers, plus a widget scaffold that renders via the MCP Apps bridge (with optional window.openai compatibility/extension hooks).
  • Reusable starter + validation: Generates a minimal Node + @modelcontextprotocol/ext-apps fallback scaffold when no upstream example fits, and outputs a repo contract validation report plus local dev and connector setup steps.

Quick Start

Use the chatgpt-apps skill to scaffold a ChatGPT Apps SDK application for an MCP-enabled tool-driven widget, using openai developer docs first and validating the result against the minimum working repo contract.

Frequently Asked Questions about chatgpt-apps

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

FAQPage Schema
How do I scaffold a ChatGPT Apps SDK project with an MCP server and widget UI?

Scaffolding a ChatGPT Apps SDK project with an MCP server and widget UI uses docs-first generation to plan tools, wire the MCP Apps bridge, and register widget resources/templates into a production-ready repository shape. It validates the result against a minimum working repo contract.

What is the best way to generate a ChatGPT Apps scaffold that matches current MCP bridge patterns?

Generating a ChatGPT Apps scaffold that matches current MCP bridge patterns requires pulling the latest Apps SDK build and quickstart guidance first, then producing tool schemas, CSP/domain metadata, and widget scaffolds that render via the MCP Apps bridge.

Can I use Node ext-apps as a fallback when scaffolding ChatGPT Apps without an upstream example?

Node ext-apps provides a minimal fallback scaffold when no upstream example fits your ChatGPT Apps workflow, generating a reusable starter project that includes tool handlers and local developer mode setup instructions.

Does the ChatGPT Apps SDK scaffold include validation for widget resource registration and tool schemas?

The ChatGPT Apps SDK scaffold includes validation for widget resource registration and tool schemas by outputting a repo contract validation report alongside the generated MCP server and widget UI code.

Why do I need docs-first generation for ChatGPT Apps MCP server wiring?

Docs-first generation for ChatGPT Apps MCP server wiring ensures that tool planning, widget resource templates, and CSP/domain metadata align with current Apps SDK conventions, preventing moving-target issues during scaffolding.

What limitations exist when generating a minimal ext-apps fallback for ChatGPT Apps?

Generating a minimal ext-apps fallback for ChatGPT Apps targets only basic Node environments using zod validation and standard node modules, meaning it may lack specialized upstream example features if your widget requires complex window.openai extension hooks.