Silphe
Your mouse has a signature as personal as your handwriting. Silphe learns it — and shows you how it moves, holds, hunts, and drifts over time.
Not whether you hit the target — how you miss it on the way there. Silphe is two things in one: a Python library that generates human-fidelity pointer movement and quantifies the movement you record, and a tiny game that captures yours while you play. Both are pure standard library. All of it is fully local.
Install
pip install silphe
No third-party dependencies. Generating and analyzing movement works on any OS; driving the real cursor is Windows-only.
Why it's interesting
Everyone clones voices; nobody clones movement. Your pointer path is as individual as a fingerprint — and far less guarded. Silphe separates two things most tools blur:
Predictive vs. reactive. Ride a smoothly drifting target and you track it with near-zero lag — you predict its path. Chase an evasive one and you're ~200 ms behind — pure human reaction time. (First real session: ~7 ms on the smooth dot, ~230 ms on the runner. Textbook.)
It drifts. Reaction, accuracy, tremor, and tracking shift with the time of day, fatigue, and the years — and Silphe plots the arc. The open question it's built to ask: when your scores climb, are you genuinely quicker, or have you just learned the board?
The games (calibration in a clown costume)
Acquire
Hit the small gold target. Fitts's law: distance × size.
Track
Follow a slowly drifting dot. Smooth pursuit, near-zero lag.
Hold
Keep dead still on a single red pixel. Your physiological tremor.
Andvari
Hunt the roach through the maze; flush it from its hiding holes. Reactive chase.
silphe-play # play silphe-arc # your fingerprint over time
Privacy
Your movement never leaves your computer. Local capture, local model, local analysis — no cloud, no telemetry, no account, no network calls. The recordings are plain files on your disk; read, move, or delete them whenever you like. Your silly walk is nobody's business but yours.