📚 Auto-generate documentation for n8n workflows with GPT and Docsify

📚 Auto-generate documentation for n8n workflows with GPT and Docsify Core Functionality This workflow transforms your n8n instance into a self-documenting system. It deploys a web server that hosts a live documentation portal, automatically generating detailed pages for each of your workflows. By analyzing a workflow's structure and configuration, it creates human-readable descriptions, visual flowcharts, and node-specific setting summaries. The system is interactive, allowing users to view, regenerate, and even manually edit the documentation through a built-in, live-preview markdown editor. Technologies Used - n8n: The automation platform hosting the workflow and serving as the web backend. The workflow also uses the n8n API to fetch data about other workflows. - Docsify.js: A JavaScript documentation site generator that runs in the browser. It fetches markdown files on the fly and renders them as a clean, navigable website. - OpenAI (GPT-4): The AI model used to analyze workflow JSON and generate descriptive text. - n8n-nodes-langchain: n8n's LangChain integration is used to structure the interaction with the AI, ensuring reliable and well-formatted output. - Mermaid.js: A diagramming and charting tool that generates visual flowcharts from text-based definitions. - HTML/CSS/JavaScript: Used to create the main documentation shell, the interactive editor, and custom client-side logic. AI Integration Analysis The AI is the centerpiece of this workflow's value, tasked with the complex job of translating raw workflow JSON into coherent documentation. 1. Input & Prompting: When a new documentation page is requested, the workflow fetches the target workflow's JSON data. This data, including all nodes and their connections, is passed to a Basic LLM Chain node. The prompt instructs a GPT-4 model to act as a technical writer, defining a strict JSON output structure with two keys: workflow_description and nodes_settings. It provides detailed formatting instructions, such as using markdown H2/H3 headers and specific formatting for node parameters. 1.