Tutorial Objectives

In 30 minutes, use Make.com to connect to ChatGPT and build an automated pipeline from "one article" to "automatic publishing on multiple platforms". This article is a complete step-by-step tutorial, which can be used even with zero coding knowledge.

What will you build?

ChatGPT summary generation module: automatically refine long articles into social media summaries
Multi-platform distribution pipeline: automatically publish to Twitter/X, LinkedIn and WordPress
Error Handling and Notification: Automatically notify you when publishing fails

Preparation list

Make.com account (free version available, includes 1000 operations/month)
ChatGPT / OpenAI API Key (obtained from platform.openai.com)
API access to at least one target platform (you can choose any one of Twitter/X, LinkedIn, and WordPress to get started)

Overall architecture

The entire workflow is broken down into 3 core modules, which can be configured in order. After everything is completed, you only need to paste the URL of an article, and the system will automatically complete the entire process from abstract generation to multi-platform publishing.

Module	Input	Output	Estimated time
HTTP trigger + content extraction	Article URL	Plain text body	10 minutes
ChatGPT summary generation	Article text	Platform adaptation summary	10 minutes
Multi-platform distribution	Summary content	Twitter/LinkedIn/WordPress Published	10 minutes

Text example image

Step 1: Create Make.com scene and configure HTTP triggering

After logging in to Make.com, click "Create a new scenario" in the upper right corner.

1.1 Add Webhook trigger

Search for Webhooks in the module selector and select "Custom Webhook" as the trigger module. Click "Add" to create a new webhook:

Configuration item	Value	Description
Data Structure	Select "JSON"	Receive standard JSON payload
Webhook name	`content-distro-webhook`	Customized name for easy identification

Once created, Make will give you a webhook URL (similar to https://hook.eu1.make.com/xxxxx). Save it - you can then just send a POST request (with {"url": "..."}) to this URL to trigger the entire workflow.

Tip: During the testing phase, you can use Postman or curl to send a test request to verify whether the webhook is working properly.

1.2 Add HTTP content extraction module

Click the + number on the right side of the Webhook, search for the HTTP module, and select "Make an HTTP request".

Configuration item	Value	Description
Method	`GET`	Get article source code
URL	`{{1.url}}`	Refers to the URL passed by the webhook
Parse Response	`否`	Get original HTML
Follow Redirects	`是`	Automatically follow 301/302 jump

This module will download the original HTML of the article. In subsequent steps, we will use ChatGPT to extract the text content and generate a summary.

Step 2: Connect to ChatGPT to generate multi-platform summaries

This is the core link of the entire workflow – using AI to replace manual social media copywriting.

2.1 Add OpenAI module

Click the + number on the right side of the HTTP module, search for OpenAI, and select "Create a Completion" (GPT-4o or GPT-4o-mini).It is recommended to use {Tool: ChatGPT/OpenAI} to complete this step, its API call cost is extremely low - it takes about 30 seconds to process an article and costs less than $0.01.

The configuration is as follows:

Configuration item	Value	Description
Model	`gpt-4o-mini`	Lowest cost, sufficient summary quality
Messages	See below	Prompt word template
Max Tokens	`1000`	Each abstract should be approximately 200-300 words
Temperature	`0.7`	Moderately creative

2.2 Prompt word template

In the User role of Messages, fill in the following prompt words (use {{2.text}} to quote the HTML content of the previous step):

code

从以下文章中提取核心信息，生成 3 种格式的社交媒体摘要：

1. Twitter 短推文（≤280 字符，含 2-3 个相关标签）
2. LinkedIn 长帖（200-300 字，专业调性，含观点和互动提问）
3. WordPress 标题 + 摘要（适合 RSS 分发，50 字以内）

文章内容：
{{2.text}}

请用 JSON 格式返回：
{
  "twitter": "...",
  "linkedin": "...",
  "wordpress_title": "...",
  "wordpress_excerpt": "..."
}

SEO+GEO：ChatGPT 生成的摘要天然包含文章核心关键词，不仅用于发帖，也可以作为后续 SEO 内容的素材。

现在，OpenAI 模块会输出结构化的 JSON，我们可以在后续步骤中直接引用每个字段。

第 3 步：配置多平台分发

将 ChatGPT 生成的摘要分发到各个平台。

3.1 Twitter/X 发布

在 OpenAI 模块右侧添加 +，搜索 Twitter，选择 "Create a Tweet"。

配置项	值	说明
Text	`{{3.text}}` 中的 `twitter` 字段	ChatGPT 生成的推文内容

首次使用时需要授权 Make 访问你的 Twitter 账号（OAuth 流程，点"授权"即可）。

3.2 LinkedIn 发布

在同一分支或新路径上，添加 LinkedIn 模块，选择 "Create a Share (Post)"。

配置项	值	说明
Author	你的 LinkedIn 个人页 ID	在 LinkedIn 设置里可以找到
Text	`{{3.text}}` 中的 `linkedin` 字段	ChatGPT 生成的专业调性帖文
Visibility	`PUBLIC`	所有人可见

3.3 WordPress 发布（可选）

添加 WordPress 模块，选择 "Create a Post"。

配置项	值	说明
Title	`{{3.text}}` 中的 `wo rdpress_title`	AI 生成的标题
Content	`{{2.text}}`	原始文章内容（经过前面提取）
Status	`draft`	先存草稿，人工审核后再发布

推荐使用 {Tool: n8n Cloud} 作为替代方案——n8n 的自托管版本更适合高频发布场景。

常见问题排查（FAQ）

Q1：OpenAI API 返回超时？

降低 Temperature 到 0.5，减少 Max Tokens 到 500。模型回复更短更快。另外检查 API Key 余额——免费额度用完后的自动续费有时会延迟生效。

Q2：Twitter 发布报错 "duplicate content"？

Twitter 不允许 24 小时内发布完全相同的推文。在提示词中加上 "{{date}}" 或文章 URL 的 Hash 值作为后缀，确保每次生成的推文文案有差异。

Q3：Make.com 免费版够用吗？

免费版每月 1000 次操作配额。如果你每天发布 5 篇文章（每篇约触发 3-5 次操作），大约可以用 60-70 天。免费版到期后，Make.com Pro 套餐（$9/月）有 10000 次操作配额，对个人创作者完全足够。

SEO+GEO：FAQ 结构满足 GEO 提取偏好，同时覆盖长尾搜索词

工具词条（触发工具悬浮卡）

正文中自然出现的工具名，平台侧会匹配已维护 tools 库生成 hover-card：ChatGPT、OpenAI、Make.com、n8n、OpenClaw、Claude

Reference video/material

Internal link guidance

Someone has successfully practiced it: 独立开发者用 n8n+OpenClaw 搭建自动化工作流，月入 5000 美元的实战案例
Recommended tools: Make.com · n8n Cloud · OpenClaw

How to build an automated content distribution workflow with Make.com + ChatGPT: From article to multi-platform publishing in 30 minutes