make-chatgpt-site
MCP server that lets MCP clients inspect, create, update, and deploy ChatGPT Sites through a local Codex installation.
README
make-chatgpt-site
Let Claude Code and other MCP clients inspect, create, update, and deploy ChatGPT Sites through a locally authenticated Codex installation.
This repository contains both layers:
- A dependency-free STDIO MCP bridge that forwards a curated Sites toolset through
codex app-server. - A
make-chatgpt-siteagent skill covering repository discovery, editing, versioning, and deployment.
Requirements
- Node.js 20 or newer
- A current Codex CLI logged in with ChatGPT
- The Sites app installed, enabled, and available to the logged-in ChatGPT account
Sites availability depends on the user's ChatGPT plan, region, and workspace settings.
Use with Claude Code
From a GitHub checkout:
claude mcp add --transport stdio chatgpt-sites -- npx --yes github:jacoblincool/make-chatgpt-site
Run /mcp in Claude Code to confirm that chatgpt-sites is connected. The bridge reuses Codex's existing ChatGPT session; it does not ask for or export browser cookies.
Consequential tools such as deployment, access changes, environment updates, and custom-domain changes require confirm=true after explicit user approval.
Create and update requests normally finish with a private deployment and a live Site URL. Ask explicitly for create-only, save-only, or no deployment when you want the workflow to stop earlier. The bridge saves versions from the pushed source commit and does not expose local archive upload.
Install the agent skill
npx skills add jacoblincool/make-chatgpt-site
The skill can be used by Codex, Claude Code, and other agents that support the common skill directory format. Install the MCP bridge separately when the agent does not already have direct access to the ChatGPT Sites connector.
Verify locally
npm test
npm run smoke
The smoke test performs only doctor and list_sites(limit=1). It does not create, update, or deploy a Site.
How it works
MCP client -> make-chatgpt-site -> codex app-server -> codex_apps -> ChatGPT Sites
The bridge checks the live connector inventory at startup and exposes only known sites.* operations. It fails closed when Sites is unavailable or an expected upstream tool is missing.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.