What problem does it solve? Qt6 QML codebases accumulate subtle defects—binding loops, delegate reuse bugs, layout conflicts, and performance anti-patterns—that manual review often misses. This Skill automates structured, read-only QML code review by combining a deterministic 47+ rule linter with six focused deep-analysis agents, producing a consolidated report with confidence-scored findings and mitigations. ## Core Features & Use Cases - Deterministic Linting: A single-pass Python script checks 47+ rules across 13 categories (imports, bindings, layout, loaders, delegates, states, images, performance, style, signals, errors, JavaScript) in under a second, with optional qmllint integration for type-level checks. - Parallel Deep Analysis: Six agents analyze bindings and properties, layout and anchoring, component lifecycle, ListView delegates, states and transitions, and performance, reporting only findings above 80/100 confidence. - Scoped Reviews: Supports both narrow diff/commit scope (review only changed lines) and wide codebase scope (glob all *.qml files in a directory). - Use Case: Before committing QML changes to a Qt6 ground station application, run this Skill to catch issues like missing required property declarations in delegates, Loader.item access without status guards, or anchors mixed with Layout.* on the same item. ## Quick Start Ask the AI to review your Qt6 QML code, for example: review the QML files in src/ui/ or check my staged QML changes before I commit.
