Can AI tag your WordPress posts without making a mess of your tags? Yes. Claude read every post, reused the tags I already had, and saved nothing until I approved it.
Tags are the part of WordPress everyone means to keep tidy and nobody does. Some of my posts had no tags at all. Others had tags that no longer matched what the post was about. Fixing that by hand means opening every post, reading it, and remembering which tags already exist. So I did it in one chat instead, with Claude connected to the site over MCP (the Model Context Protocol) and two free plugins. Watch it on YouTube.
1
Chat to review every post
2
Free plugins from WordPress.org
4
Abilities doing the work
0
Changes saved before approval
Two free plugins to tag WordPress posts with AI
Both are free on WordPress.org. One decides what your site can do; the other connects it to your assistant.
01
Gives Claude abilities
AcrossAI Abilities Manager
Registers the operations your site can perform, including reading posts and managing taxonomy terms, with one admin page to switch each one on or off. See every ability.
02
Connects Claude
AcrossAI MCP Manager
Publishes those abilities to your AI assistant over MCP, authenticating as a real WordPress user with that user’s capabilities and nothing more.
The abilities doing the work
Read
content/list-posts + content/get-post
Find the posts and read what each one is actually about, along with the tags it already has.
Check
taxonomies/list-terms
Pull every tag that already exists on the site. This is the step that keeps things clean.
Create · Only when needed
taxonomies/create-term
Add a new tag only when nothing existing fits.
Save
taxonomies/assign-cpt-terms
Save the tags to each post, either appending to what’s there or replacing it, so you decide whether existing tags survive.
The best part isn’t the speed
Any tool can stamp tags onto posts quickly. The hard part is not ending up with WordPress, wordpress-tips and WP as three separate tags that each hold two posts.
01 · Read
Read before tagging
It read each post for what it’s actually about, not just its title, and looked at the tags it already had.
02 · Reuse
Existing tags first
Because it checked the full tag list first, it reused what I already had instead of inventing near-duplicates.
03 · Explain
New tags come with a reason
It only proposed a new tag when it could say why none of the current ones fit.
04 · Approval
Nothing saved until I said so
I saw the full plan, post by post, before anything changed. When a suggestion was off, I said so in plain words and it adjusted.
05 · Control
Append or replace, my choice
Tags could be added to a post’s existing ones or replace them, so I decided what survived.
That’s the difference between an assistant and a bulk script you run and hope for the best with.
The prompt I used to tag WordPress posts with AI
What I asked
“Go through my published posts and find the ones with no tags or tags that don’t fit the content. Check my existing tags first and reuse them wherever possible. Show me the proposed tags for each post and don’t save anything until I approve.”
What Claude did
It listed the published posts, read each one, and pulled the site’s full tag list. Then it showed me a plan, post by post, reusing existing tags wherever they fit and flagging the few new ones with a reason. Only after I approved did it save the tags to each post.
Works with whichever assistant you already use
MCP is an open standard. Claude, ChatGPT, Gemini, Cursor, Grok and other MCP-capable clients connect to the same site with no changes on the WordPress side. And because your site is the MCP server, your content goes straight from WordPress to the assistant, with no third-party relay in between.
Verify after every write
The habit that makes this trustworthy
Once the tags are saved, ask Claude to read a few posts back and confirm the tags match what you approved. Then glance at Posts → Tags to check that no stray duplicates slipped in. It takes seconds, and it’s your independent check that what you were told actually happened.
How to tag your own WordPress posts with AI
Step 01
Install the two free plugins
Add MCP Manager and Abilities Manager from WordPress.org.
Step 02
Connect your assistant
Run the Quick Connect wizard, generate an Application Password, and paste the ready-made config into Claude or your MCP client.
Step 03
Ask for a plan, not changes
Use the prompt above: check existing tags first, and don’t save anything until you approve.
Step 04
Review and adjust
Go through the plan post by post. If a tag is off, say so in plain words and let it revise.
Step 05
Approve, then verify
Approve the plan, then have Claude read a few posts back and check Posts → Tags for duplicates.
Frequently asked questions
Can AI tag WordPress posts automatically?
Yes. Connected over MCP, an assistant like Claude can read each post, suggest tags from what it’s actually about and save them — across the whole site in one conversation.
How do you stop AI from creating duplicate tags?
Have it read your existing tags first. taxonomies/list-terms pulls every tag on the site, so the assistant reuses what you have and only creates a new tag when nothing existing fits.
Will it overwrite the tags a post already has?
Only if you want it to. taxonomies/assign-cpt-terms can either append to a post’s tags or replace them, so you decide whether existing tags survive.
Does anything change before I approve it?
No. Ask for the full plan first and nothing is saved until you approve it. Afterwards, ask the assistant to read a few posts back to confirm the tags match.
Clean up your tags in one chat
Install the two free plugins, connect your assistant, and ask for a tagging plan before anything changes.
Watch it: Tagging WordPress posts with AI · Learn more: the AcrossAI platform · every plugin integration