building-chatgpt-apps

Build interactive ChatGPT Apps with MCP servers and embedded widgets.

Updated Jan 13, 2026
One-click install
npx skills add https://github.com/hafiznaveedchuhan-ctrl/hackaton2pages --skill building-chatgpt-apps-hafiznaveedchuhan-ctrl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: building-chatgpt-apps
Source: https://github.com/hafiznaveedchuhan-ctrl/hackaton2pages/tree/main/.claude/skills/building-chatgpt-apps
Command: npx skills add https://github.com/hafiznaveedchuhan-ctrl/hackaton2pages --skill building-chatgpt-apps-hafiznaveedchuhan-ctrl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Designing and deploying ChatGPT Apps with rich widgets can be complex and error-prone; this skill provides a structured guide to assemble MCP servers, widget HTML, and proper response metadata for reliable integration.

Core Features & Use Cases

  • Widget architecture and MCP server setup using FastMCP
  • Accurate response metadata injection via _meta["openai.com/widget"]
  • Developer Mode setup and testing workflow for ChatGPT Apps
  • Real-world use cases include building chat-assisted dashboards, task automations, and content explorers

Quick Start

Create an MCP-enabled widget app, expose it publicly, and register it in ChatGPT Developer Mode to begin testing.

Frequently Asked Questions about building-chatgpt-apps

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

FAQPage Schema
How do I build interactive widgets for ChatGPT Apps using MCP servers?

You can build interactive widgets for ChatGPT Apps by combining FastMCP-based MCP servers with embedded HTML, enforcing the text/html+skybridge MIME type and openai.com/widget response metadata to render visual UIs.

What response metadata is needed to render custom widgets in ChatGPT?

ChatGPT requires accurate response metadata injected via _meta["openai.com/widget"] to render custom widgets, enforcing the text/html+skybridge MIME type for reliable visual UI integration.

How do I test ChatGPT widgets in Developer Mode?

To test ChatGPT widgets in Developer Mode, create an MCP-enabled widget app, expose it publicly, and register it within the ChatGPT Developer Mode environment to begin testing workflows.

Can I use FastMCP to set up interactive UIs for ChatGPT Apps?

Yes, FastMCP provides the pipeline to set up MCP servers and interactive UIs, enabling window.openai integrations like sendFollowUpMessage, toolOutput, and callTool for ChatGPT Apps.

Why are my ChatGPT App widgets not displaying properly?

ChatGPT App widgets fail to display properly when the text/html+skybridge MIME type or the _meta["openai.com/widget"] response metadata is missing or incorrectly configured in your MCP server pipeline.

What real-world use cases work best for ChatGPT widgets with MCP?

Real-world use cases for ChatGPT widgets with MCP include building chat-assisted dashboards, task automations, and content explorers that leverage interactive visual UIs and developer-mode workflows.