OpenAI Codex: How To Install n8n MCP Server
May 6, 2026 · Mike Murphy
In this tutorial, you will learn how to set up the n8n instance-level MCP server in OpenAI Codex. Once installed, Codex can build n8n workflows for you from simple text prompts, with no copying or pasting of JSON between apps.
Description
In this tutorial, you will learn how to set up the n8n instance-level MCP server in OpenAI Codex. Once installed, Codex can build n8n workflows for you from simple text prompts, with no copying or pasting of JSON between apps.
🧠 What You Will Learn
- n8n MCP: enable the instance-level MCP server
- Version check: confirm n8n 2.18.3 or later
- Codex setup: add an MCP server in settings
- Streamable HTTP: configure the URL and headers
- Bearer tokens: what they are and how to format them
- Build workflows: prompt Codex to create n8n workflows
- Workflow access: enable or remove MCP per workflow
🔗 Links
🌐 Hostinger VPS (Affiliate Link):
https://www.hostg.xyz/SHIDN
OpenAI Codex:
https://chatgpt.com/codex/
⏱️ Timestamps:
00:00 Intro: About The Tutorial
00:39 n8n Version Number (2.18.3 or Later)
01:14 Enable n8n MCP
02:01 MCP Connection Details
02:23 Codex: Add MCP Server
03:16 Headers (Key + Value)
04:05 Configuration JSON
05:09 Bearer Token
06:24 Bearer (Access Token)
07:19 Save MCP Setup
07:47 Codex MCP Server Settings
08:05 Quit & Restart Codex
08:30 Confirm Codex Can Access MCP
08:57 Codex: List Workflows
09:16 Codex Build N8n Workflow
10:24 Test Codex n8n Workflow
12:03 n8n Instance-Level MCP Access