Two claps and Claude Code opens, VS Code starts, the music plays. A 2-second startup ritual — no clicking, no typing. Ready for Windows and Mac, installed in 5 minutes.
Built and used every day by the TX Lab + PatronoX. What you see here is what runs on our machine right now.
Clap detection sounds simple until you open the first tutorial. Python, sounddevice, numpy, FFT, cosine similarity, threshold calibration, microphone permission on Mac, audio driver on Windows — and that is just the beginning.
The TX Lab team spent days debugging each system. This kit is the result of that, packaged so you can install it in 5 minutes.
Kit Jarvis delivers everything solved: tested code (Windows + Mac), visual installation and customisation guides, and prompts ready to paste into any AI for a personal instructor that stays with you until it works.
Any code on GitHub assumes you already know what you are doing. Kit Jarvis assumes the opposite.
Along with the scripts comes an installation prompt you paste into Claude, ChatGPT or Gemini. From there the AI becomes your co-pilot: one step at a time, waiting for your confirmation, fixing errors as they appear. It does not move to the next step until the current one works.
Infinite support, no queue, no schedule, no question limit.
In the .zip you download:
Mac version, with Terminal and AppleScript already integrated. No portability patches.
Ready script for Windows. Commented in English. You open it, read it, understand it and change whatever you want.
Double click and everything installs itself. You do not touch the terminal even once during installation.
Step by step, from zero to the first "two claps and it worked". Every command, every permission explained.
Change the music, the app, the voice command, the language, the number of claps and the sensitivity. Each one in under 1 minute.
3-step quick start for those who already know what they are doing.
Paste it into Claude, ChatGPT, Gemini — the AI becomes your instructor and guides you step by step, fixes bugs, answers questions. No queue, no schedule, no question limit.
Same for customisation. One adjustment at a time, testing with you.
Most clap detection projects online use amplitude — if the sound is loud, it counts. That detects keyboard, table knock, door closing. It does not work.
Kit Jarvis uses spectral analysis. During calibration, it records the shape of your clap sound — like a sonic fingerprint. Then, in real time, it compares each sound against that profile using cosine similarity.
Yes. The kit comes with installation prompts ready to paste into Claude, ChatGPT or any AI. The AI takes the role of instructor: one step at a time, confirms it worked before continuing, fixes bugs immediately. You do not need to understand what you are doing — just follow along.
Windows 10/11 and macOS. Requires Python (the guide teaches installation from scratch) and a microphone — the laptop mic works. Clap detection runs 100% offline. The optional voice command requires internet.
All of that is editable. The customisation guide covers: which app opens, which music plays, which terminal command runs, the voice activation phrase, the language, the number of claps and the detector sensitivity. Each adjustment takes under 1 minute.
The two AI prompts included in the kit are the support. Paste them into Claude or ChatGPT and you have a specialist available 24h, no queue, no question limit, no expiry.
Instant download after payment. You receive a .zip with all files: Python scripts for Windows and Mac, automatic installer, two step-by-step PDFs and two AI prompts. One-time payment of €4.90 — no subscription, no access platform, no upsell. Yours forever.
From zero to a working clap-activated AI assistant. Everything you need is inside the kit — scripts, guides, and AI prompts to walk you through each step.
If you have never touched code, the kit includes a guide that teaches you to install Python from scratch — step by step, with screenshots. Takes about 5 minutes. The AI prompt handles any errors automatically.
python --version → 3.11+
A single script installs all dependencies: sounddevice, numpy, scipy. It detects your operating system and adjusts automatically. Windows or Mac — same command, same result.
python install_jarvis.py
The script records your clap 5 times and creates a sonic fingerprint unique to you. This is what makes it ignore keyboard sounds, table knocks, and background noise — it only responds to your clap.
python calibrate.py
Edit a single configuration file to define what happens on two claps: which app opens, which music plays, which terminal command runs. Everything is commented — takes under 2 minutes.
config.json → customize everything
Run the listener script. From this moment, two claps activate your complete setup. Claude Code opens, VS Code launches, the music plays. You went from idea to Jarvis in under 30 minutes.
python jarvis.py
Scripts ready for Windows and Mac. Visual installation and customisation guides. AI prompts to walk you from zero to your first clap. €4.90, one time.