Spuncin
Your posts
are training AI.
For free.
Spuncin shuts the door on the crawlers that copy your work to train models,
and leaves it open to the ones that bring you readers.
The problem
They read everything. Nothing comes back.
Companies run automated crawlers across blogs to collect text for their models. They don't pay for it, they don't credit you, and they don't send a single person to read the original. Your hosting serves them without telling you.
Blocked by default
Always allowed through
The question everyone asks
And Google? Google stays.
Spuncin doesn't touch search engines. It isn't a setting you can get wrong, because the setting doesn't exist: whoever can send you a visitor is never blocked, at any level.
Your ranking doesn't move, your links keep their preview image, your feed keeps working.
How it works
Install it, and the job is done.
No account, no subscription, nothing to study. It starts working the moment you activate it.
Install the plugin
Look for Spuncin under Plugins in your WordPress dashboard and install it like any other. Nothing else is asked of you.
Choose how strict to be
Three levels, each one telling you plainly what it costs. The one already selected costs nothing.
Forget about it
Spuncin recognises the crawler and turns it away before it reads a single line. You can look at how many were blocked, if you feel like it.
Three levels
You decide where the line goes.
Try it here. Pick a level and watch who gets blocked.
Blocks whoever copies to train models
The crawlers that collect your posts to build datasets and train models. They take, and nothing comes back to you.
Also blocks AI search
Assistants that read your site to answer a question, and that can quote you with a link back.
Blocks everything, people included
Requests made because a real person has just asked an assistant to open your page.
Privacy
It works without knowing anything about you.
Spuncin never contacts a server. Not ours, not anyone's. Everything happens inside your site, which is why there is nothing to add to your privacy policy.
- No external callsThe plugin never phones home. Not for list updates, not for statistics, not for anything.
- No personal dataIt counts how many crawlers were stopped each day and their names. No addresses, no visitors, nothing that points back to a person.
- Nothing to declareNo third party is involved, so there is no data processor to name and no cookie banner to add.
- Nothing to payFree, open source, and no paid version waiting behind the feature you actually need.
For the curious
The technical part, if you want it.
Everything below is optional. The plugin behaves the same whether you read it or not.
How does it recognise a crawler?
By the name it announces in its user agent, matched as a whole word and not as a piece of text. Looking for "Spider" inside a user agent would also catch Baiduspider, Sogou and Screaming Frog, so fifteen names too generic to be safe are never matched at all. Search engines are checked further with a reverse DNS lookup, so nobody walks in by simply claiming to be Googlebot.
Why does it write robots.txt as well?
Because the two most valuable opt-outs cannot be caught any other way. Google-Extended and Applebot-Extended have no user agent of their own: Google crawls with its usual crawler and reads that name only from robots.txt. Refusing them there is the only thing that works, and it is what lets you leave AI training without losing a single visitor.
What does a blocked crawler get?
A 403 and a short page that names neither the crawler nor the plugin. Telling whoever is knocking which rule stopped them only makes it easier to work around it.
Does it slow the site down?
No. On a normal visit the work is one comparison against a list already held in memory. There is no database query, no external call, and nothing that breaks page caching.
What if my site is behind Cloudflare?
Blocking still works. Only the reverse DNS check on search engines needs the real visitor address, which a CDN replaces with its own. There is a filter, spuncin_client_ip, for whoever runs that kind of setup.
Where this is going
Saying no is half the job.
Spuncin comes out of a project that won an IOTA hackathon, where the starting idea was a larger one: not just shutting the door, but putting a toll on it.
Today your blog can do one thing, refuse. Sooner or later it should be able to say something else: come in, but what you take has a price. A price you set, collected without you having to negotiate with anyone.
We are working on it. We are not racing to be first, because when money is involved something rushed is worse than something late. It ships when it is safe for whoever collects and whoever pays, and when it does what it says.
Close the door.
It takes a minute.
Free, in your language, and no account to create.
Install the plugin free, on WordPress.orgRequires WordPress 6.4 or newer and PHP 7.4 or newer. It installs from your dashboard, like any other plugin.