What to Type for AI

AI Character Prompt Library for Indie Game Devs

Stop losing hours to inconsistent AI character art. This library gives you 40 structured, tested prompts across 5 game archetypes — Warrior, Rogue, Mage, Tank, and Ranged Support — each with a neutral idle, dynamic action pose, defeated state, and gear variant. Every prompt is formatted for Midjourney, DALL-E 3, Stable Diffusion, and Leonardo.ai in their correct syntax, not adapted from a generic template.

The system works because every prompt is built from five explicit layers: a character base brief that seeds identity, an anatomy anchor that locks proportions, a clothing layer stack written innermost-to-outermost to prevent mesh merging, a geometrically described expression chain that beats vague emotion words, and a style token set tuned for your tool.

Included: 6 expression portrait prompts (angry, fearful, happy, shocked, confused, neutral) that inherit any archetype's identity for UI cards and dialogue art. Plus a full troubleshooting guide for AI drift, wrong limb counts, blurry pixel art, and gender expression issues — with specific fixes, not vague advice.

One structured session. One cohesive character library. No artist required.

Who It Is For

Solo indie game developers and pixel or 3D artists who need a complete character visual library without commissioning or hiring a character artist.

What Is Inside

• 40 fully structured AI prompts across 5 archetypes: Warrior, Rogue, Mage, Tank, Ranged Support — 4 prompts per archetype covering idle, action, defeated, and gear-variant states
• Every prompt written in 4 tool-specific versions: Midjourney, DALL-E 3, Stable Diffusion, and Leonardo.ai — with correct syntax for each
• 5-layer prompt anatomy system: base brief, anatomy anchor, clothing layer stack, expression chain, and style token set — explained and labeled inside every prompt
• 4 complete style token sets for pixel art, anime, 3D game-ready, and semi-realistic — each with consistency lock instructions and tool-specific settings
• 6 expression-only bonus prompts (neutral, angry, fearful, happy, confused, shocked) that slot onto any archetype character for UI portraits and dialogue cards
• Anchor method for cross-image consistency — how to keep the same character face across 20+ generations without paying for Midjourney's highest plan
• Anti-drift troubleshooting section: 8 specific problems with tested fixes, including clothing merge prevention, limb anatomy anchoring, and pixel art blur prevention
• Full session walkthrough — Warrior in pixel art style, start to finish, with every SD setting value, full prompt text, and upscaling instructions filled in

Use It When

• Generating a complete sprite sheet of 8+ character states for your 2D RPG protagonist in a single 2-hour session using only Stable Diffusion
• Producing matching idle, action, and UI portrait assets for all five character classes in your game without visual drift between them
• Switching an entire character set from pixel art style to semi-realistic painted style for a different game project by swapping only the style token layer

Related

AI Design Prompt Formulas →Prompt formulas for brand identity, client briefs, design revisions, and AI-assisted creative workflows. How to Lock Brand Colors and Boilerplate in AI Designs →Keep hex codes, brand palettes, and locked boilerplate consistent across AI-generated designs instead of drifting on every regeneration. How to Prevent Your Approved Logo from Morphing During Brand Guideline Updates →Keep your approved logo stable while iterating other brand elements. The revision-control prompt structure that stops unwanted AI drift in logo design. How to Stop Endless Client Revisions on AI-Generated Design Work →End revision cycles that kill your margin. Use prompt formulas to translate vague feedback into regenerations that land first try. Brand Identity Prompt Formulas →Turn any 3–4 sentence client brief into an on-brand logo concept, color palette, and brand guideline visuals in 2–4 AI generations — not 15+.... Prompt Formulas for Client Revisions →Most revision cycles aren't caused by bad prompts — they're caused by a missing translation layer between what clients feel and what prompts can specify....