Turn content tagging into a hands free job. The flow watches an RSS feed, uses AI to suggest tags, checks your WordPress site for matches, and adds missing tags before publishing. It is built for content and SEO teams that publish often and want clean, consistent tags.

Here is how it works. A schedule checks the RSS feed every minute. An AI model generates a structured list of tags, and an auto fixing parser keeps the format clean. The flow pulls all current WordPress tags through the API, normalizes case and slugs in a code step, and finds what is missing. It creates only the tags you need, fetches the updated list, filters the matches, and aggregates tag IDs. A WordPress node then posts with the correct tags. It loops through many articles in batches and can also be called as a subworkflow.

Setup needs an OpenAI key and WordPress access to the REST API. Expect faster publishing and fewer errors. It fits blogs, news rooms, and B2B content hubs. Make sure your slug and name casing rules match your site to avoid duplicates. If you run it as a subworkflow, enable Run Once for Each Item.

• RSS feed polling every minute to capture new articles
• AI chat model generates a clean list of suggested tags
• Structured parser enforces a tags array and fixes format issues
• WordPress API calls read existing tags and create missing ones
• Code step normalizes case and slug style to match site rules
• If branch only creates tags when they do not exist
• Aggregate step collects tag IDs for the final post
• WordPress node publishes a post with title, content, and tags
• Split out loops through each missing tag to create it
• Can run as a subworkflow for other content pipelines

• Reduce manual tagging from 20 minutes to 1 minute per post
• Automate up to 90 percent of tagging work with AI
• Improve tag accuracy by enforcing consistent case rules
• Handle 10 times more posts with batch processing
• Connect OpenAI and WordPress in one reliable flow

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI and WordPress. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, create an OpenAI credential. Choose API Key, paste your key from the OpenAI account page, name the credential clearly, and save.
4. In the n8n credentials manager, create a WordPress credential. Double click any WordPress or HTTP Request node pointing to WordPress, select Create new credential, then follow the on screen steps. Use your site URL and an account with permission to manage and create tags. Application Passwords work well for many sites.
5. Open the RSS Feed Trigger node and paste your feed URL. Keep the every minute schedule or adjust to your needs. Execute the node once to verify items are fetched.
6. Open the AI tag generation nodes and review the prompt and the structured schema that expects a tags array. Keep the schema simple to ensure clean output.
7. Open the HTTP Request nodes that point to the WordPress tags endpoint and confirm the base URL matches your site. Test the GET tags call to confirm you see existing tags.
8. Check the code step that builds missing tags. Confirm the slug format you want. If your site uses a different case rule, update the replace and case functions accordingly.
9. Run the workflow in test mode. Watch for the If branch result. If missing tags are found, the flow will create them, fetch the updated list, and then aggregate tag IDs.
10. Open the final WordPress node and review the title and content fields. Map any additional fields you need. Execute this node alone to confirm a post is created with tags.
11. If you call this as a subworkflow, open the Execute Workflow settings and enable Run Once for Each Item. This ensures each article is processed correctly.
12. Troubleshooting: If tags do not create, check WordPress user permissions and REST API access. If duplicates appear, align slug casing rules. If the AI output is empty, review your prompt and test with a single item.