What problem does it solve? Choosing a language, framework, or platform for every new project from scratch leads to inconsistent stacks and zero accumulated expertise. This Skill reads your cumulative tech preference matrix and recommends choices aligned with your prior projects, balancing performance, update ease, and learning-cost recovery. ## Core Features & Use Cases - Preference Matrix Lookup: Reads your tech matrix (languages, runtimes, frameworks, deploy targets, DB, state management, testing, lint) from a personal wiki entity file and matches new project requirements against it. - Drift Detection: Scans existing repositories (package.json, requirements.txt, pyproject.toml) to report which projects have drifted from your declared stack, then lets you fix the drift or update the matrix. - Matrix Updates: When you deliberately adopt a new technology, updates the preference matrix and logs the change with a one-line rationale. - Use Case: You ask "what stack should I use for this new project?" and receive a recommendation scored on performance, update ease, learning-cost recovery, ecosystem stability, and Korea-market fit—defaulting to your existing matrix when scores tie. ## Quick Start Ask the assistant to recommend a stack for your new project consistent with your previous projects, and it will audit your preference matrix and report any drift.