(if you know how to use Claude)
I'm so sick of seeing a new bloated trading journal pop up every single week, charging $40 to $60 a month for what's basically an Excel template with a dark mode UI and a win-rate counter. We have Claude now, we have vibe-coding, and if you have even basic prompting skills there is zero reason to keep paying monthly rent to these SaaS guys.
You can build a system for free that actually understands your strategy, better than any generic tool out there. I got so annoyed with those subscriptions that I just coded my own native setup using mainly Claude, kept adding stuff I actually needed for my own charts, and ended up wiring biometrics like heart rate tracking into it just to see how hard my pulse spikes when I'm FOMOing or hesitating on execution.
Why this works now
The vibe-coding hype isn't just noise, it's a window. You can describe what you want in plain English, Claude writes the structure, you test it on your own trades, and you iterate until it fits your brain. No subscription model, no feature bloat you'll never use, no dashboard designed for everyone which means it's optimized for nobody.
Most trading journals ask you to fill in how you felt, on a scale from one to five or with tags like "confident" and "anxious". That's testimony from the same brain that wanted the trade in the first place, which makes it evidence of nothing. When I went back through my own notes from earlier setups, the impulsive ones almost never got labelled impulsive at the time. They got labelled "slightly early entry on a valid setup," which tells you everything about how useless real-time emotional data actually is.
What I built
I started with a basic prompt structure for logging trades, then added rules, then integrated Apple Watch data because I wanted to see my actual heart rate three minutes before I hit the button. Turns out your BPM doesn't lie, even when your journal entry does. No FOMO? Well, your BPM was 98 and climbing, so yes you did.
That's the layer most journals miss, because they can't access it without building something native. Your journal shows what you did, your biometrics show who you were. That gap is where most trading losses actually live, and it's measurable if you bother to measure it.
Two routes from here
If you're tech-savvy, go build your own with Claude. Seriously, it's way better than paying $250 a year for static software. You'll learn more about your own edge by building the system, which means the construction process itself teaches you more than six months of filling in someone else's form ever could.
But if you don't want to deal with coding, API keys, or hosting setup yourself, I'm opening up a small pilot for the tool I built. I refuse to run a greedy $50/month subscription, so I'm letting a handful of traders test it and give feedback. You get biometrics integrated, you get the setup done, and if you join the pilot you lock in a 50% lifetime discount.
We're looking for 10 people who trade at least three times a week, already journal in some form, and are willing to sit down for two short interviews during the six-week pilot. That's it, no upsells, no tiers with features held hostage.
If you want in on the pilot, go to stoplossdivision.com and join the waitlist. If you just want to know how I structured the prompt logic to build your own for free, reach out and I'll walk you through it. Either way, stop paying rent on a glorified spreadsheet when you have Claude sitting right there.
Disclaimer: This trading journal is for educational purposes only, forms no financial advice, and the author is in no way liable for any trading losses. By using this material, you acknowledge that all trading decisions remain your own responsibility.