Getting oriented#

New here — human or AI agent? Start with the runtime helpers, then dig into the topic pages.

import multicolorfits as mcf
mcf.overview()              # mental model, conventions, task index
mcf.recipes('cutout')       # copy-paste recipe matching a keyword
cat = mcf.overview(as_dict=True)   # structured catalog for tools

The same catalog is published as machine-readable files at the docs site root:

Both files are generated from multicolorfits/_overview.py by scripts/make_llms_txt.py. tests/test_overview.py fails the CI if either file drifts from the catalog — regenerate, never hand-edit.

multicolorfits.overview(query=None, *, as_dict=False)[source]#

Print an orientation to multicolorfits — the colorize-then-combine model, conventions, and a task→function index.

Parameters:

  • query (str, optional) – If given, defer to recipes() (print matching runnable recipes).

  • as_dict (bool) – Return the structured catalog for tools / programmatic use instead of printing: {'layer_first', 'conventions', 'recipes': [{'task', 'category', 'functions', 'code', 'notes'}, ...]}.

Return type:

Any

Notes

Designed as the first call an agent (or newcomer) makes: import multicolorfits as mcf; mcf.overview().

Examples

>>> import multicolorfits as mcf
>>> mcf.overview()                 
>>> mcf.overview('cutout')         
>>> cat = mcf.overview(as_dict=True)
>>> sorted(cat)
['conventions', 'layer_first', 'recipes']
multicolorfits.recipes(query=None)[source]#

Print runnable task→code recipes.

With no argument, list every recipe (task + category). With a query keyword (a task / function / topic), print the matching recipes’ copy-paste code.

Examples

>>> import multicolorfits as mcf
>>> mcf.recipes()              # the full menu   
>>> mcf.recipes('cutout')      # transparent stamps   
>>> mcf.recipes('lab')         # perceptual compositing
Parameters:

query (str | None)

Return type:

None