Skip to content

Gentleman-Programming/gentle-shell / skills/tui-design

gentle-ai-tui-design

Verified

Trigger: terminal UI design, TUI redesign, Gentle Shell look, IDE in a terminal, sidebar/status bar polish, critique TUI. Refine terminal UIs with visual proof.

FileKindSize
NOTICE.mdmarkdown403 B
SKILL.mdskill md3.3 KB
references/commands.mdmarkdown3.1 KB
references/critique.mdmarkdown2.1 KB
references/design-doc.mdmarkdown1.4 KB
references/terminal-craft.mdmarkdown3.9 KB
references/visual-loop.mdmarkdown1.8 KB

references/critique.md

View on GitHub
# Critique (scored)

Judge from a screenshot, not from code. Write the design assessment first, then check measurable evidence (overflow, contrast, render counts) so the evidence does not anchor the judgment.

## Health score

Score each heuristic 0-4 (0 broken, 2 workable, 4 excellent); mark `n/a` and renormalize to /40 when one does not apply.

1. Visibility of system status (live work, progress, mode, permissions).
2. Match with developer expectations (IDE and terminal conventions).
3. User control (collapse, scroll, escape, undo of view changes).
4. Consistency (glyphs, roles, section shapes, terms).
5. Error prevention (dangerous modes visible, confirmations where needed).
6. Recognition over recall (hints where the action is, palette discoverability).
7. Efficiency for experts (keyboard paths, density, no wasted rows).
8. Minimalist design (each fact once, no decoration).
9. Error recovery (messages with cause and next step).
10. Help and documentation (discoverable keys and commands).

Bands: 36-40 excellent, 28-35 good, 20-27 acceptable, 12-19 poor, under 12 critical.

## Cognitive load check

Single focus, chunks of 4 or fewer, clear grouping, visible hierarchy, one decision at a time, 4 or fewer options per decision, no need to remember across screens, progressive disclosure. 0-1 failures is low load; 4 or more is critical.

## Personas

- Power user: lives on the keyboard, wants density and zero repetition.
- Keyboard/accessibility: contrast, focus visibility, no color-only state.
- Stress tester: long sessions, 50 agents, 200 todos, narrow terminal.

## Severity

- P0 blocking: the task cannot be done or the UI breaks (rail disappears, content unreadable).
- P1 major: significant difficulty or contrast failure. Tie-breaker: would a user report it? Then at least P1.
- P2 minor: a workaround exists.
- P3 polish.

## Report shape

1. Target, sizes captured, PNG paths.
2. Health score table and total.
3. Specificity verdict: could any other TUI ship this unchanged?
4. What works (2-3).
5. Priority issues (3-5): `[P?] what / why it matters / fix / command`.
6. Persona red flags.
7. Minor observations.
8. 2-4 questions with concrete options, last.
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall