Pattern reading machine
Your input. It's read. Your map.
How it works
Your input
Set a baseline — where things stand right now. Add to it over time. The more honestly you put in, the more clearly it reads back.
The machine reads
It works through your material. It counts what appears independently. It finds what is consistent, what conflicts, what is building, and what is absent. It does not generate, guess, or fill gaps.
Your output
A map of what it found — bounded entirely by what you wrote. Every claim carries a confidence label. Nothing is invented. Nothing is added.
You stay in control
If the map misreads something, you correct it. Corrections govern every subsequent reading. The machine does not override you.
What it reads
The same reading engine works across any domain where something is moving, stuck, or under load. Pick the cartridge for your context — each one carries the domain knowledge specific to that situation.
What this is — and isn't
It is
A reading and sorting tool
It is not
An AI generating responses
It is
Evidence-bound — reads only what you put in
It is not
A scoring or assessment tool
It is
Private — nothing leaves your device
It is not
A decision-maker — you stay in control
It is
A single file — no account, no server
It is not
A substitute for your own judgement
17 domains. One reading engine. One file per map.
No account. No server. Works offline.