Teaching views
Install Murali Kit and you get themes, named colors, and lesson diagrams on top of the engine.
python3 -m pip install murali-kit==0.3.0
from murali_engine import Scene, Label
from murali_kit.colors import WHITE
from murali_kit.composite import TitleCard
from murali_kit.themes import DarkTheme, apply_theme
scene = apply_theme(Scene(), DarkTheme())
TitleCard("Murali Kit", "Teaching views on Murali Engine").add_to_scene(scene)
scene.add(Label("Hello", height=0.3, color=WHITE))
scene.preview()
This is the usual authoring path. You still import primitives from murali_engine. Kit does not
replace Scene or Timeline.
Themes and colors​
from murali_kit.themes import DarkTheme, LightTheme, apply_theme
from murali_kit.colors import WHITE, BLUE_D, GOLD_C
theme = LightTheme()
scene = apply_theme(Scene(), theme)
apply_theme styles an existing scene and returns it. Themes do not construct scenes. The palette
table is on Space and color.
Composite​
from murali_kit.composite import TitleCard, Opening, MuraliLogo, ChatInputBox
Title cards, openings, and the logo are authored compositions. Opening can also live inside a
SceneView when the ident needs its own camera.
AI​
from murali_kit.ai import (
AttentionMatrix,
ContextWindow,
KvCacheView,
NeuralNetworkDiagram,
NextTokenDistribution,
TokenSequence,
TransformerBlockDiagram,
TensorView,
)
These are lesson diagrams, not a model runtime. The category roadmap is AI Visualization.
Maths​
from murali_kit.maths import FunctionGraph, NumberLine, VectorField, StreamLines
Linear-algebra teaching views also live under murali_kit.maths (LabeledVector2D,
MatrixVectorFlow, and related). Engine Axes / NumberPlane stay on murali_engine when you
need the generic renderable.
Story and layout​
from murali_kit.storytelling import Stepwise
from murali_kit.layout import Group
Stepwise is the guided-reveal helper. Group / stacks share transforms without a second clock.
Examples​
python examples/hello_shapes.py
python examples/motion_basics.py
python examples/title_card.py
python examples/neural_networks.py
Catalog: murali-kit/examples.
Kit currently depends on murali-engine>=0.3.0,<0.4.0.