本文へ移動
cccskills
無料GitHub で公開

game-ui-workflow

Design and implement game HUDs, menus, icons and portrait systems from visual proposals through reusable engine assets and input checks. Use when a game interface needs a coherent visual and behavior contract.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.4 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Game UI workflow

Inspect the current game camera, input rules, UI framework and representative screens. Reuse the project's selected visual language and framework. Establish the information and actions the player needs before generating decorative art.

Design against real gameplay

Use a native game capture as the context for a proposal. Keep it intact when comparing HUD alternatives, so changing the background cannot disguise contrast or obstruction. Define shared surface, type, spacing and selection tokens; preserve semantic colours for ownership, health, danger and accepted choices.

Separate visual art from live data. Decide which values come from runtime state, which actions dispatch existing commands and what remains unimplemented. Show unavailable data honestly instead of inventing a currency, timer or gameplay rule. Include hover, selected, disabled, empty and overflow states when relevant.

Review the actual interface scale and aspect ratios. Corner anchors, minimum sizes, long labels and compact rosters should remain usable at a narrow viewport. Test the intended selection/grouping rules; unit type and team may be independent keys. A combined health display and a range of differing stats are different data.

Prepare reusable assets

Keep approved art and source masters, with stable asset IDs and a manifest of source paths/hashes. Export icons at their intended display scale. Check alpha edges on light/dark backgrounds, shared optical weight and silhouette readability. For stretchable panels, identify border regions explicitly; verify corners at several sizes before choosing nine-slicing or another rendering approach.

Static and animated portraits need the same crop and a reliable still fallback. Budget concurrent video decoders/render targets, share identical playback when appropriate and release hidden views. Portrait art should be data-driven rather than adding a new UI branch for each character. Preserve GUIDs and existing bindings when updating imported assets.

Integrate and prove interaction

Use the engine's existing UI tools or an available editor bridge. State the required framework/package instead of assuming a particular MCP is installed. Keep layout and world-input exclusion in agreement: a click over a panel must not also command the world. Verify hover/click, drag, keyboard focus, selection, minimap behaviour and pause/return paths that the changed screen owns.

Read current values and cooldowns from their authoritative systems. Avoid duplicating combat/production timers in presentation code. Reuse views, bound expensive updates and avoid per-frame texture readback for a minimap.

Save the real assets and check them in a fresh engine session. Deliver the theme, asset mapping, native captures, tested interactions and open decisions. A browser mockup approval is a design decision, not a claim of engine integration.

Example request: "Give the existing HUD a compact unit panel and six matching command icons." Success includes consistent icons, real selection data, working commands, overflow behaviour and no click-through. Dependencies: the current UI sources, game captures and target-engine access for implementation.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Choose a workflow for 3D concepts, meshes, materials, animation, game VFX, UI, levels, audio and native visual review.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

Rig and animate supplied characters in Blender, revise motion from references, and deliver editable sources plus a verified game export.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

Prepare strict-front character sheets, isolated parts and optional multiview references for 3D modeling from an approved image.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

Generate and edit images, create video, audio, 3D assets or material maps through fal.ai MCP or the Python queue client. Use for fal model discovery, reference uploads, generation, recovery and local delivery.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

Prepare a precise handoff for an agent or developer implementing a feature, continuing a task or reviewing a finished change.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

Integrate approved character clips into a game's movement and combat systems, including turn compensation, gait phase, event timing and interruptions. Use after animation authoring or to repair runtime transitions.

日本語の概要は準備中です。原文の説明を表示しています。

witnesstodark/mr-mak-workspace3612026年10月11日 更新

witnesstodark のスキルをすべて見る

このスキルの問題を報告する