designing-art-prompts-and-technique-cards

Designs art and UI image prompts, technique cards, and pre-approval visual review workflows.

Updated Jul 9, 2026
One-click install
npx skills add https://github.com/alsdmlals4-eng/Base --skill designing-art-prompts-and-technique-cards-alsdmlals4-eng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: designing-art-prompts-and-technique-cards
Source: https://github.com/alsdmlals4-eng/Base/tree/main/skills/designing-art-prompts-and-technique-cards
Command: npx skills add https://github.com/alsdmlals4-eng/Base --skill designing-art-prompts-and-technique-cards-alsdmlals4-eng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Game and creative projects often generate images without a clear consumer, approval state, or identity constraints, producing decorative assets that drift from project canon. This Skill turns visual requests into explicit production contracts with defined consumers, protected identity axes, review gates, and readback requirements. ## Core Features & Use Cases - Prompt Contract Design: Builds structured image generation and edit prompts covering consumer, canon references, protected identity, composition, resolution, and engine consumption constraints. - Visual Requirement and Coverage Gates: Routes production information to text/table/DB outputs, blocks generation without an actual consumer, and prevents coverage gaps from auto-triggering image batches. - Candidate Review and Reuse Harvest: Compares generated candidates against canon and task contracts, manages status lifecycle from GENERATED_EXPLORATION to PROJECT_ASSET_APPROVED, and classifies reusable visual value. - Use Case: When asked to create a character expression variant, the Skill locks identity constraints, applies the conversation approval gate, generates one candidate, reviews it against approved references, and records readback before any promotion. ## Quick Start Ask the assistant to design an image generation prompt for a specific game screen or character asset, providing the project relation, approved visual references, and the intended consumer surface.

Frequently Asked Questions about designing-art-prompts-and-technique-cards

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write an image generation prompt for a game character?

Define the actual consumer first, then lock protected identity axes like face proportions, costume, and palette before specifying the changed expression or pose. The Skill structures prompts with canon references, composition, resolution, and forbidden drift elements.

When should I not generate an image for project documentation?

Skip image generation when the need is production information like system explanations, worldbuilding, or relationship diagrams. Route those to text, tables, Mermaid, or databases instead, since decorative documentation images are classified as DO_NOT_GENERATE.

Can generated images be promoted directly to approved project assets?

No, generated results are candidates only. They must pass visual QA against canon, receive explicit user decision, and complete readback verification before reaching PROJECT_ASSET_APPROVED status, with runtime evidence required for final verification.

How does the Skill handle external reference images and copyright?

External sources are recorded as REFERENCE or BENCHMARK entries with provenance and rights status. The Skill extracts functional principles into a reference brief while forbidding replication of identifiable expression, and blocks release when rights are unverified.

What is the difference between a coverage gap and image generation approval?

Coverage preflight only detects missing visual assets like button states or telegraphs; it never grants generation permission. Each image still requires the Visual Requirement Gate selection and explicit conversation approval before creation.