Essays • AI and engineering • • 5 min read

I Wrote a Linter That Strips the AI Tells From My Own Writing

A shell script catches every em dash on this blog. The real AI tells live in register, where no regex can see them. What I lint for, what I read aloud for, and why a blocklist deletes real voice.

A handwritten page on a dark desk under a lamp, with a red pencil lying across it

The em dash is the easy one.

I can catch every em dash I will ever type with a short shell script, and I do, on every commit. The script walks my published posts and the book chapters and fails the commit if it finds a real em dash, the YAML-escaped version of one, a spaced double hyphen that the markdown renderer would turn into a dash, or the space-less word--word that slips past me. Then it runs a second time after the build, against the rendered HTML, because the sneaky ones do not come from prose. They come from a title in the frontmatter or a string in a template, and the only place you can be sure to catch them is the page a reader actually sees.

That gate works. It has never once let an em dash through since I wired it in. And it taught me the thing this whole post is about. The em dash was never really an AI tell, more a typographic habit, and the real tells are harder. They do not survive contact with a regular expression.

The ones a script can catch

A few of them are mechanical enough to grep for, so I do.

Figurative "lands" and "resonates," the ones that show up as "the idea lands" or "that really resonated." A literal rocket can land and a literal string can resonate, so the script cannot just ban the words, but it can flag them for me to look at. "The part nobody talks about," in every costume, which is a move that fakes a secret out of a common observation. And the opener I caught myself using on essay after essay: "I have spent fifteen years doing X, and here is how I use AI." Other AI-engineering blogs open that exact way too, and so did a run of my own essays.

So I grep for the strings. The first time I did, I learned something. The moment the script catches one, the writer reaches for the synonym. Swap "resonated" for "stuck with me," swap one credential opener for another, and the sentence still reads like a machine wrote it. The generated feeling survives the swap because it was living in the register the word arrived in.

The ones it cannot

There is no grep for register. A sentence can be technically correct and completely lifeless. A paragraph can resolve a little too neatly. Generated prose has a particular evenness to it when nothing in it was ever uncertain, and no pattern I know how to write sees that.

So the second layer is a judgment, and the rule I use for it is older than any of this. Read the thing aloud against a sample of my own older writing, and if it does not sound like me, rewrite it until it does. I keep a body of my freewriting as the reference and the standard is written down so it cannot quietly drift. Whether a thing sounds like me is a question I have to answer myself, or hand to something that has read enough of me to stand in.

A blocklist cannot do that job. Some of the words on every generic list are genuinely mine. I use "unleash" and "palpable" and "realm" in their old physical sense, the way I have used them for years, and a literal-minded pass that strips them in the name of de-robotizing my prose just deletes my actual voice. What disqualifies a word is the register it showed up in, the generic figurative one or the real one, and no script knows the difference.

The system leaks, on purpose

None of this is a finished detector. I find new tells after they have already shipped, on pages that were live for weeks. When I find one I add it to the gate if it is mechanical and to the written standard if it is not, and lately I have started pointing an agent at my own published archive and asking it where I still sound like a machine. It finds things. Some of them I argue with. The list is not done and it is not going to be done, because the writing keeps maturing and the tells move with it.

That is the same loop I run on everything. When a failure gets past me I fix the one instance and then build the thing that catches the next one, knowing I am never fully ahead of it. I wrote about that habit in the engineering context in Recursive DevOps. It turns out prose has the same shape.

If you write with AI, this is what I have. Two different things, and only one of them can be automated. A gate for the mechanical tells, the dashes and the dead phrases, which a script handles forever once it is written. And a taste oracle for the rest, a real sense of your own register, supplied by you or trained into something from a corpus of you. The gate I can hand over. The taste oracle is the writing itself. The rest of how I work this way lives at AI-assisted engineering.