comprehension-debt

Comprehension-driven development: Claude quizzes you on your own PR diff so you catch the gap between what you think your code does and what it actually does — before you merge.

Measure the code nobody understands — verified-understanding loop for AI-era codebases (scan · gate · repay · prove)

A CI/CD-integrated cognitive control plane that gates pull requests behind interactive architectural quizzes to prevent developer deskilling and rubber-stamping of AI-generated code.

Turn your Claude Code sessions into a plain-English project diary, living docs, and a visual map. One command, zero API keys.

You merged 4,200 lines of AI code last month. How much does your team actually understand?

Always-active Claude Code plugin: Three Axes Framework — calibrates AI behavior across Mastery, Consequence, and Intent axes to prevent comprehension debt.

Hold an AI agent's summary against what it actually changed. Zero dependencies, one file, MIT.

Adaptive programming instructor for Claude Code — structured courses, discovery-first teaching, and pluggable curricula

Audit whether your codebase is as real as it claims. A zero-dependency static analyzer that grades every module claimed-only → reachable → tested → asserted from the code itself — never commit messages — and measures the AI-slop tax. No execution, no install of the code under audit.