Step 2: Connect the Blog Feed Parsing Pipeline

Fetch the blog homepage HTML, extract post blocks, and parse titles and links for each post.

1. In Retrieve Blog Feed, set URL to https://blog.n8n.io/.
2. In Extract Feed HTML, set Operation to extractHtmlContent and extract posts using CSS selector article.item with Return Array enabled.
3. In Distribute Posts, set Field to Split Out to posts and Destination Field Name to post.
4. In Parse Title Link Tags, set Data Property Name to post and extract link, title, and tags using selectors .item-title a and .item-tags a.

Step 3: Build Article URLs and Filter Recent Posts

Compose each article URL, fetch article metadata, and ensure only recent posts move forward.

1. In Compose Article URL, add an assignment for articleUrl with value =https://blog.n8n.io{{ $json.link }} and keep Include Other Fields enabled.
2. In Retrieve Article HTML, set URL to ={{ $json.articleUrl }}.
3. In Get Publish Timestamp, extract published from selector meta[property="article:published_time"] with Return Value set to attribute content.
4. In Recent Publish Check, set the boolean condition to ={{ (new Date($('Scheduled Run Trigger').item.json.timestamp) - new Date($json.published)) / (1000 * 60 * 60) <= 480 }} to allow only posts from the last 480 hours.
5. In Fetch Clean Article Text, set URL to =https://r.jina.ai/{{ $('Compose Article URL').item.json.articleUrl }}.

Step 4: Set Up AI Summarization and Drafting

Use OpenAI to summarize each article and draft a LinkedIn post based on the extracted content.

1. Open OpenAI Chat Engine and select the model gpt-4o-mini.
2. In Generate Article Summary, set Text to ={{ $json.data }} and define the prompt message that asks for a markdown summary.
3. Credential Required: Connect your OpenAI API credentials in OpenAI Chat Engine. This node provides the language model for Generate Article Summary.
4. In Initialize Draft Prompt, set prompt to You will be passed a scraped article from a company blog. I'm an influencer for their product and will write a LinkedIn post for every news that is coming out. Please write an engaging post based on the article.
5. Open OpenAI Chat Engine B and select the model gpt-4o-mini.
6. In Draft LinkedIn Update, set Text to ={{ $('Fetch Clean Article Text').item.json.data }} and set the prompt message to ={{ $json.prompt }}.
7. Credential Required: Connect your OpenAI API credentials in OpenAI Chat Engine B. This node provides the language model for Draft LinkedIn Update.
8. Use Iterate Items Batch to control batching between Generate Article Summary and Initialize Draft Prompt for each item.

Step 5: Configure Human Review and Publishing

Send the draft for human approval, route the outcome, and publish approved posts to LinkedIn.

1. In Human Review Gate, map the review fields to include link, summary, postDraft, and timestamp using the existing expressions like ={{ $('Generate Article Summary').item.json.text }} and ={{ new Date($('Get Publish Timestamp').item.json.published).toDateTime().toRelative() }}.
2. Keep updateForReviewId set to ={{ $('Prepare Retry Prompt').isExecuted ? $('Prepare Retry Prompt').item.json.reviewToUpdate : null }} to support review retries.
3. In Route Review Outcome, keep the switch rules matching approved, rejected, and type=chat for retry handling.
4. In Prepare Retry Prompt, set prompt to ={{ $json.messages[0].content }} and reviewToUpdate to ={{ $json.reviewId }}.
5. In Publish LinkedIn Post, set Text to ={{ $json.responseValues.postDraft.value }}, Title to ={{ $('Parse Title Link Tags').item.json.title }}, and Original URL to ={{ $('Compose Article URL').item.json.articleUrl }}, with shareMediaCategory set to ARTICLE.
6. Ensure Publish LinkedIn Post connects back to Iterate Items Batch to continue processing remaining posts.
7. Credential Required: Connect your GoToHuman credentials in Human Review Gate.
8. Credential Required: Connect your LinkedIn credentials in Publish LinkedIn Post.

Step 6: Test and Activate Your Workflow

Run a manual test to confirm the feed parsing, AI outputs, review gate, and LinkedIn publishing all work correctly.

1. Click Execute Workflow and verify Retrieve Blog Feed returns HTML and Extract Feed HTML returns an array of posts.
2. Confirm Compose Article URL generates valid URLs and Get Publish Timestamp returns a published value.
3. Check that Generate Article Summary produces a markdown summary and Draft LinkedIn Update produces a post draft.
4. Approve a draft in Human Review Gate and confirm Publish LinkedIn Post publishes successfully to LinkedIn.
5. When everything works as expected, toggle Active to enable scheduled runs.