Conciseness

The body is dense and assumes Claude's competence — no space is spent explaining what a TUI is or how frameworks work — but the '2026-09-04' date stamps appear twice outside any deprecated/old-patterns section, and mouse-interaction guidance recurs across 'Make It Feel Modern and Clickable', 'Input, Focus, and Editing', and two Anti-Patterns rows. These minor instances could be trimmed, matching the 'efficient; minor instances of over-explanation' anchor rather than the fully lean anchor above it.

Actionability

For an instruction-only design skill the guidance is fully actionable: 'prefer rounded box-drawing corners (╭ ╮ ╰ ╯) over square (┌ ┐ └ ┘)', 'one cell of padding inside panels', 'Do not mix EventStream with read/poll', 'WCAG's 4.5:1 normal-text contrast', '80x24 is a useful test case', plus a concrete test matrix and named reference apps (lazygit, k9s, yazi). Per the rubric's code-vs-instruction note, absence of code is not penalized when guidance is this specific, and it is not below the anchor since common cases are covered.

Workflow Clarity

The sections form a coherent progression — 'Choose the Layout from the Work' → interaction/inputs → 'State and Async Work' → capability policy → accessibility → rendering/lifecycle → 'Verification That Changes Confidence' — with concrete validation checkpoints ('inspect at least one real PTY session', 'verify the shell cursor and echo behavior are restored'). It is a design-concern reference rather than a sequenced build procedure, so it sits at 'clear sequence with most checkpoints present' rather than the fully explicit step/feedback-loop anchor.

Progressive Disclosure

References are well-signaled and one level deep — 'Read [app-patterns.md](references/app-patterns.md) for examples and tradeoffs' and 'Read [visual-catalog.md](references/visual-catalog.md) when selecting glyphs' — and both files exist. However, the 176-line body inlines reference-grade material (full Terminal Capability Policy, verification matrix, anti-pattern tables) that could live in references/, so it matches 'good structure; most content appropriately placed; minor organization gaps' rather than the ideal clear-overview split.

Specificity

It lists multiple concrete capabilities — 'designing terminal user interfaces, creating TUI layouts, choosing TUI color schemes, implementing keyboard AND mouse navigation, building terminal dashboards' — comprehensively covering the skill's domain with no vague filler, matching the 'lists multiple specific concrete actions; comprehensive coverage' anchor. It is not the 4-anchor because there is no meaningful coverage gap.

Completeness

Both halves are explicit: what ('designing terminal user interfaces... building terminal dashboards, or working with any TUI framework') and when ('Use this skill when...', 'Activates on mentions of...', 'Always consult this skill when a TUI is being built or restyled, even if the user only asks for it to "look modern"'). This matches the anchor for clearly answering both what and when with concrete trigger phrases.

Trigger Term Quality

Trigger terms are comprehensive and natural, spanning the whole domain and its synonyms: 'TUI design, terminal UI, Ratatui layout, Ink components, Textual widgets, Bubbletea views, terminal color palette, keybinding design, panel layout, split panes, terminal dashboard, box-drawing characters, sparklines, progress bars, modal dialogs, focus management, mouse support'. Named frameworks give the synonym coverage the 5-anchor requires.

Distinctiveness Conflict Risk

The TUI niche is distinct and framework-named triggers are unambiguous, but 'terminal dashboards', 'sparklines', and 'progress bars' as bare activation terms carry minor overlap risk with charting/dataviz skills, fitting 'mostly distinct; minor overlap risk with closely related skills'. It is not 5 because those triggers are not scoped to a TUI context, and not 3 because the overlap is confined to adjacent visualization skills rather than many similar skills.