Staging Practical Tooling in a K-12 Python Curriculum: A Cognitive-Load and Dependency-Graph Analysis

Introductory Python curricula must decide when to introduce the surrounding tool chain — interpreter install, REPL, virtual environments, package managers, IDEs, version control, literate notebooks, test runners. The standard choices are just-in-case (full tool chain upfront) or just-in-time (tools introduced when a felt need surfaces). We argue on cognitiveload grounds that just-in-case tooling at entry imposes extraneous load that displaces language-schema acquisition, and that the just-in-time alternative requires deliberate dependency-aware staging rather than ad-hoc scatter. We analyse nine tool categories through two lenses — element interactivity (Sweller) and a pedagogical dependency graph — and propose a six-tier staging mapped against a four-volume K-12 Python curriculum whose endpoint is competitive programming. The design locates a three-to-five page minimal-tooling chapter at Vol 1 Ch.1 (install + REPL + one-file run) and a full deep-tooling chapter at Vol 2 Ch.6 (venv + pip + IDE + git local + Jupyter awareness). Vol 3 adds pytest habituation, git-remote supplementary material, and timeit informally; Vol 4 adds contest-environment descaling and profiling. A companion "Python Toolbox" reference document absorbs platform drift that otherwise churns print volumes. The design is corroborated by convergent practice in three major Python introductory corpora (Runestone, PY4E, Zelle) and by Software Carpentry's separate-lesson evidence for shell and git.