Step 2: Connect Google Sheets

Google Sheets is used to both look up previously processed queries and log published articles.

1. Open Lookup Sheet Rows and set Document to [YOUR_ID] and Sheet to gid=0.
2. Set the lookup filter query to {{ /*n8n-auto-generated-fromAI-override*/ $fromAI('values0_Value', ``, 'string') }}.
3. Credential Required: Connect your googleSheetsOAuth2Api credentials for Lookup Sheet Rows. This is an AI tool for Content Query Agent, so ensure credentials are configured on the parent tool connection.
4. Open Append Sheet Log and set Document to [YOUR_ID] and Sheet to gid=0.
5. Verify the append mappings use expressions such as {{ $json.article.slug }} and {{ $('Format Publishing Payload').item.json.article.focusKeyphrase }}.
6. Credential Required: Connect your googleSheetsOAuth2Api credentials for Append Sheet Log.

Step 3: Set Up Content Discovery and Search

This section generates a unique search query and pulls recent articles from Google Custom Search.

1. Open Content Query Agent and confirm the Prompt text is set to the multi-line query instruction block.
2. Ensure OpenAI Chat Engine is connected as the language model for Content Query Agent and uses model gpt-4-turbo-preview.
3. Credential Required: Connect your openAiApi credentials for OpenAI Chat Engine.
4. Open Fetch Recent Articles and set URL to https://www.googleapis.com/customsearch/v1.
5. Configure query parameters: key to [CONFIGURE_YOUR_API_KEY], cx to [YOUR_ID], q to {{ $json.output }}, dateRestrict to d7, and num to 5.
6. Confirm the flow order: Scheduled Daily Trigger → Content Query Agent → Fetch Recent Articles → Select Top Article → Filter Blocked Sources.

Step 4: Set Up Article Generation and Formatting

These nodes generate the original article and normalize the JSON payload for publishing.

1. Open Compose Original Article and set Model to gpt-4-turbo-preview, Max Tokens to 4096, and Temperature to 0.7.
2. Review the message template to confirm it includes the expressions like {{ $json.content }}, {{ $json.title }}, and {{ $json.source }}.
3. Credential Required: Connect your openAiApi credentials for Compose Original Article.
4. Open Format Publishing Payload and keep the provided JavaScript Code that parses and formats the response into publishing, SEO, and metadata fields.
5. Verify the flow order: Filter Blocked Sources → Compose Original Article → Format Publishing Payload.

Step 5: Configure Publishing and Notifications

The final stage posts the formatted article to your CMS, logs results to Google Sheets, and sends a Slack update.

1. Open Post to CMS Endpoint and set URL to webhook_url and Method to POST.
2. Set JSON Body to the full expression payload beginning with { "title": {{ JSON.stringify($json.article.title) }}, "slug": {{ JSON.stringify($json.publishing.slug) }}, ... }.
3. Confirm the request sends Content-Type header as application/json.
4. Open Append Sheet Log and confirm it appends after Post to CMS Endpoint.
5. Open Dispatch Slack Update and set Channel to [YOUR_ID] and Text to the templated message using expressions like {{ $('Post to CMS Endpoint').item.json.article.title }}.
6. Credential Required: Connect your slackOAuth2Api credentials for Dispatch Slack Update.

Step 6: Test and Activate Your Workflow

Run a full test to validate the content pipeline before turning on daily publishing.

1. Click Execute Workflow and verify Content Query Agent outputs a short search phrase.
2. Confirm Fetch Recent Articles returns results, Select Top Article parses them, and Filter Blocked Sources outputs at least one valid article.
3. Check Compose Original Article outputs a JSON-formatted article and Format Publishing Payload builds publishing fields correctly.
4. Verify Post to CMS Endpoint returns a successful response, then confirm Append Sheet Log writes a new row and Dispatch Slack Update posts in Slack.
5. When satisfied, switch the workflow to Active so Scheduled Daily Trigger runs it every morning.