Mindket

Summary box added to the top of each WordPress post

Newversion 1 · updated

Overview

An n8n workflow that has OpenAI write a short key-points box and places it above the text of your WordPress posts. It skips posts already done, logs each one in Google Sheets and tells a Slack channel.

Summary box added to the top of each WordPress post is an n8n workflow in the Writing category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.

At a glance

Runs in
  • n8n Cloud: import the file into a new workflow there
  • Your own n8n server: the same file and steps
  • WordPress: the posts are read and changed through the site’s REST API
  • Google Sheets: the list of posts already done, one row each

Platform

Connects 4 apps

How it works

Trigger

Manual run

It starts when you press Test workflow, or from a schedule or a webhook once you switch one on. A webhook call carries:

  • The ID of the post, as post_id

Actions

  1. WordPress returns your newest posts, five at a time
  2. Posts already listed in the Google Sheet are skipped
  3. Each other post is fetched in full and its HTML turned into Markdown
  4. A classifier skips posts that already hold an AI summary
  5. OpenAI writes a short box of key points, in the post’s language
  6. WordPress puts the box above the post’s content and keeps its excerpt
  7. The post is added to the sheet, and a notice goes to Slack

Result

Each new post carries a summary box at the top, with a row in your sheet and a notice in your Slack channel

Integrations

What you need before you start.

OpenAI

Language models

Runs gpt-4o-mini to write each summary box, and a second model to check whether a post already has one.

NeedsOpenAI API key

WordPress

Posts

Returns the posts, then updates each one with the summary above its content.

NeedsA WordPress API credential for your site

Google Sheets

Log of finished posts

Tells the workflow which posts it has done, and takes one row for each new one.

NeedsA Google service account credential, and a sheet with five columns

Slack

Notices

Posts a message with the post’s link and edit link when a post is updated.

NeedsA Slack OAuth2 credential, and a channel

Workflow check

Checked by Mindket on 13 Nov 2026

Steps
14 steps, starting by hand
Not reached
6 nodes no run reaches: Webhook, Schedule Trigger, Set fields - From Webhook input, Date & Time - Substract, WordPress - Get Last Posts and 1 more
AI
2 steps call an AI model: Text Classifier and OpenAI
Sign-ins
Signs in to: WordPress, Google service account, Slack and OpenAI
API keys
No API key is left in the file
Step types
Mindket’s checker reads all 12 of its step types

Read from the workflow’s file, not from a run.

About

What it is

This n8n workflow adds an AI-written summary box to the top of your WordPress posts and keeps a log of what it did. It fetches posts from your site, skips any that a Google Sheet already lists, and turns each post’s HTML into Markdown. A classifier checks that no summary is in the text yet. OpenAI then writes a box of bullet points in the post’s language, and a call to the WordPress REST API puts it above the old content. Each finished post is added to the sheet and announced in Slack. A manual run starts it, and a schedule and a webhook are included, switched off. The file holds 32 nodes, eleven of them sticky notes.

Who it suits

Bloggers, magazines and content teams on WordPress who want a short summary on top of long posts without a plugin. Skip it if each summary needs your approval before it shows. As imported, a post on your live site changes as soon as OpenAI writes the box.

Who uses it, and how
IndustryWho uses itHow they use itExample
PublishingMagazine editorsAdd a key-points box to long articlesA tech magazine puts four bullet points above each long review.
EducationCourse site ownersPut a recap above each lesson pageA language school adds a recap box to every grammar lesson.
E-commerceContent managersHelp skimmers find the point of a buying guideA bike shop adds a box to its guide on city bike frames.
NonprofitsCommunications teamsGive each report post a short summaryA local charity adds one to every annual report post.

You need solid n8n for this one: four credentials, a sheet with set columns, your domain in two nodes and a careful first test on a few posts.

Where to use it

Any n8n that can reach your WordPress site, OpenAI, Google and Slack will run it. The summary goes into the post itself, and the log goes into the sheet.

  • n8n Cloud: import the file into a new workflow there
  • Your own n8n server: the same file and steps
  • WordPress: the posts are read and changed through the site’s REST API
  • Google Sheets: the list of posts already done, one row each
  • Slack: one notice in a channel for each post

Setup

Download saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Make the sheet first, with its five column names in the first row.

What you need first
NeedDetail
n8nOn n8n Cloud or on a server of your own
WordPressA WordPress API credential for your site, and a test copy to try it on
OpenAIAn API key. The summary step uses gpt-4o-mini
GoogleA service account credential, and a Sheet shared with its email address
SlackAn OAuth2 credential and a channel for the notices
The sheet’s first row
ColumnWhat it holds
post_idThe post’s ID. The workflow looks it up to skip posts it has done
summaryThe HTML box OpenAI wrote
post_linkThe post’s address
edit_linkThe address of the post in your admin
summary_dateWhen the summary was added
  1. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
  2. Make the Google Sheet with the five columns above.
  3. Add a WordPress API credential and select it in the four WordPress nodes and in the HTTP node that updates the post.
  4. In the update node and in the Set node that prepares the sheet and Slack fields, replace <your-domain.com> with your site’s domain.
  5. Add an OpenAI credential and select it in OpenAI and OpenAI Chat Model.
  6. Add a Google service account credential, select it in both Google Sheets nodes, and choose your sheet in each.
  7. Add a Slack OAuth2 credential, select it in the Slack node and choose a channel.
  8. Open the system message of OpenAI. It describes an electric vehicle site, so name your own theme and replace its Good and Bad examples.
  9. Run it from When clicking ‘Test workflow’ on a test site first. It takes five posts.
  10. For production, switch on Schedule Trigger, or Webhook with a Header Auth credential, then press Publish.

How to use it

Press Test workflow once. It takes your newest five posts, writes a box for each, and logs and announces them. For steady use, a schedule checks for new posts, or your site calls the webhook with a post_id when a post goes live.

What to change for a different result
To changeWhere
The summary’s theme, language and lookThe system message of OpenAI
How many posts a run takesThe Limit in the two WordPress nodes that list posts, 5 by default
How often it checksSchedule Trigger, and the 30 seconds in the date node after it
Which postsFilters, such as a category, in the WordPress nodes
The Slack noticeThe text in the Slack node

Example

A bike shop’s blog has a long guide on choosing a city bike frame. The webhook gets its post_id when the post goes live.

The workflow finds no row for that ID, reads the post and writes a four-bullet box on frame materials, sizes and weight. WordPress then shows the box above the guide. The sheet gets a row, and Slack gets a notice with the post’s link and edit link.

For newcomers and experts

If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. This file has three ways to start and one shared path. Run the manual start first. Then switch on one of the other two for production, and leave the other off.

For experts. The sheet and the classifier both guard against a second summary on the same post.

Where to look closer
NodeWhat it doesWhat to know
The sheet lookupLooks up the post’s post_idIt always outputs an item, so a post with no row goes on
Text ClassifierSends a post on if no AI summary is in its textPosts that already have one go back to Loop Over Items
The update nodePOSTs the new content and the old excerpt to the postIt uses the WordPress credential and your domain
Schedule TriggerEvery 30 seconds as it standsThe date node after it takes 30 seconds from its timestamp for the date filter

Good to know

It edits live posts, and two of its models are on OpenAI’s shutdown list.

  • Models: OpenAI calls gpt-4o-mini. OpenAI’s deprecations page, read on 12 November 2026, lists it for shutdown on 14 January 2027, with gpt-6-luna as the replacement
  • Classifier model: OpenAI Chat Model names none, so n8n’s default applies, gpt-5-mini. The page lists it for shutdown on 11 December 2026, with gpt-5.6-terra as the replacement
  • Live posts: the update writes the summary and the old content back as the new content. Back up your site and try a few posts first
  • Sheet columns: auto-map matches by name, and the workflow’s date field is summary_date. The node’s saved column list shows summarized_date, so name your column summary_date
  • Slack: the first run can post a notice for every post it summarizes. Switch the node off for a large first batch
  • Schedule: it is set to every 30 seconds, n8n’s default for seconds. Pick a calmer interval
  • Your data: each post’s text goes to OpenAI
  • Credentials: the JSON holds none. Mindket joined two trigger branches that ended without a next step

README and files

README

Summary box added to the top of each WordPress post

An n8n workflow that puts a short AI-written summary above the text of your WordPress posts, with no plugin on the site.
It fetches posts, skips any it has done before, has OpenAI write a box of key points, updates the post through WordPress's
REST API, logs it in a Google Sheet and tells a Slack channel.

What is inside

  • workflow.json: the workflow as n8n exports it, 32 nodes. Eleven are sticky notes that explain its steps. It holds no
    credential.
  • README.md: these notes.

What we changed from the original copy

  • Two of the three triggers led nowhere. After Schedule Trigger the chain stopped at WordPress - Get Last Posts, and
    after Webhook it stopped at WordPress - Get Post1. Both nodes now lead into Loop Over Items, which the author's
    notes say is fine for one post as well. The Schedule and Webhook triggers are still switched off.
  • Its author's Google Sheet ID and Slack channel ID were in the sheet and Slack nodes. Both are cleared, so choose yours.
  • Two sticky notes told you to copy a Google Sheets template of the author's. They now list the columns to create.
  • Every credential link is gone, so select your own in each node that needs one.
  • Nothing else. Its expressions needed no repair.

Before you start

  • An n8n instance, on n8n Cloud or your own server.
  • A WordPress site with its REST API open to a WordPress API credential in n8n, and a site you can test on first.
  • An OpenAI API key.
  • A Google service account credential in n8n, and a Google Sheet shared with that account's email address.
  • A Slack account with an OAuth2 credential in n8n, and a channel for the notices.

Set it up

  1. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
  2. Make a Google Sheet with this first row: post_id, summary, post_link, edit_link, summary_date.
  3. Add a WordPress API credential and select it in WordPress - Get All Posts, WordPress - Get Post2,
    WordPress - Get Last Posts, WordPress - Get Post1 and Wordpress - Update Post.
  4. In Wordpress - Update Post and Set fields - Prepare data for Gsheets & Slack, replace <your-domain.com> with your
    site's domain.
  5. Add an OpenAI credential and select it in OpenAI and OpenAI Chat Model.
  6. Add a Google service account credential and select it in Google Sheets - Get rows and Google Sheets - Add Row. Choose
    your sheet in both.
  7. Add a Slack OAuth2 credential, select it in Slack - Notify Channel and choose a channel.
  8. Open the system message of OpenAI. It describes an electric vehicle site. Name your own theme and replace the Good and
    Bad examples.
  9. Run it from When clicking ‘Test workflow’ on a test site or a few drafts first. It takes five posts.
  10. For production, switch on one trigger: Schedule Trigger, or Webhook with a Header Auth credential, which expects
    your site to post post_id in a JSON body. Then press Publish.

How a post runs

  1. When clicking ‘Test workflow’ starts the first run, and WordPress - Get All Posts returns your newest posts, five by
    default. Schedule Trigger and Date & Time - Substract feed WordPress - Get Last Posts, the posts after the last
    run. Webhook and Set fields - From Webhook Input feed WordPress - Get Post1, one post by its ID.
  2. Loop Over Items takes one post at a time.
  3. Google Sheets - Get rows looks for the post's post_id, and If skips the post when a row exists.
  4. WordPress - Get Post2 fetches the raw post, and HTML to Markdown converts its content.
  5. Text Classifier checks whether the text already holds an AI summary. If so, the post is skipped.
  6. OpenAI writes the summary as an HTML block with bullet points, in the post's language.
  7. Wordpress - Update Post sends the block plus the original content back to the post, and keeps its excerpt.
  8. Set fields - Prepare data for Gsheets & Slack gathers the post's ID, title and links, the summary and the date.
    Google Sheets - Add Row logs them, and Slack - Notify Channel posts a notice.

Change the result

  • The summary's wording, theme, language and HTML: the system message of OpenAI.
  • How many posts a run takes: the Limit in WordPress - Get All Posts and WordPress - Get Last Posts, 5 by default.
  • How often it checks: the interval in Schedule Trigger, and the 30 seconds in Date & Time - Substract. Change both
    together.
  • Which posts: add filters, such as a category, to the WordPress nodes.
  • The notice: the text in Slack - Notify Channel.

Good to know

  • OpenAI calls gpt-4o-mini. OpenAI's deprecations page, read on 12 November 2026, lists it for shutdown on 14 January
    2027, with gpt-6-luna as its replacement.
  • OpenAI Chat Model names no model in the file, so the node uses its own default, gpt-5-mini (n8n's repository, read the
    same day), although a sticky note says gpt-4o-mini. The deprecations page lists gpt-5-mini for shutdown on
    11 December 2026, with gpt-5.6-terra as its replacement.
  • It edits live posts. Wordpress - Update Post writes the summary and the old content back as the new content, so back up
    your site and try a few posts first.
  • Auto-map in Google Sheets - Add Row matches fields to columns by name. The workflow's date field is summary_date, but
    the node's saved column list still shows summarized_date. Name your column summary_date.
  • The first run can post a Slack notice for every post it summarizes. Switch Slack - Notify Channel off for a big first
    batch.
  • Schedule Trigger is set to every 30 seconds, n8n's default for seconds. Pick a calmer interval first.
  • The post's text, with its HTML turned into Markdown, goes to OpenAI.
  • OpenAI bills one classifier call for each post and one summary call for each post that needs it.
Files
NOTICE.md  668 B
README.md  6 KB
workflow.json  39 KB

3 files, 45 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-12

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

anaksor-ocr

Sokheng · Kit

Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…

New

Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…

New

ACE-Step 1.5 XL SFT

ACE-Step · Kit

Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…

New

Similar in Writing

See all in Writing

Rewrites English text with fewer grammar mistakes, using a T5 model trained on the JFLEG dataset. Made by vennify for writers, students and developers who want…

New

Translates between 33 languages on your own device with a 1.8B model squeezed to 1.25 bits, about 440 MB as GGUF files for llama.cpp. For developers adding off…

New

Summarizes, answers questions about and continues book-length text with a fine-tuned 27B model that uses Gated DeltaNet layers, in English and Japanese. For pe…

New