Intent

Beta

The co-processor that tells the brain where.

From "click Save" to a target. Command understanding.

Platforms	Python
Engine	FlyCore v0.1.0
Weights	v0.1.0-beta
Input ports	Instruction embedding → olfactory (2,282); Screen → visual (11,390)
Output ports	Target embedding → Cursor / Keys ← central
Co-processors	Vision-language model

The 3D view needs WebGPU or WebGL2.

Install

Python

pip

pip install flycore

Usage

Intent turns a short instruction plus a screen into a target embedding the control weights can act on. It is the bridge between language and motion, and it ships as a co-processor for Cursor and Keys.

Python

import flycore

brain = flycore.load("flylabs/intent")      # engine + weights + adapters
cursor = flycore.load("flylabs/cursor", coprocessors=[brain])
cursor.do("click Save")
# target grounded by Intent, motion by Cursor

Figures

Figures marked target are design targets for the first release, not measurements.

Grounding	94%target ScreenSpot desktop
Co-processor	380 MBtarget on disk, 4-bit

Limits

Beta: English instructions.
Requires Cursor or Keys to act; Intent alone only produces targets.