MCPInstall

agent-sprite-forge-2d-game-assets

by reason-machinesMedia & Creative

Generate 2D sprite sheets, animated GIFs, transparent PNGs, layered maps, and engine-ready prototypes for Godot/Unity using Codex-driven asset pipelines.

Installs
217
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/reason-machines/ai-agent-skills --skill agent-sprite-forge-2d-game-assets
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads agent-sprite-forge-2d-game-assets automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Agent Sprite Forge — 2D Game Asset Generation

Skill by ara.so — AI Agent Skills collection.

Agent Sprite Forge is a Codex-first 2D game asset workflow where the agent decides the plan, image generation creates the raw visuals, and deterministic scripts turn those visuals into reusable game assets. This skill enables AI agents to generate sprite sheets, layered maps, transparent PNGs, animated GIFs, and engine-ready exports for Godot and Unity.

What It Does

  • Sprite Sheets: Characters, monsters, props, attacks, spells, projectiles, impacts, idles, walks, and reference-guided variants
  • Layered Maps: Ground-only bases, dressed references, prop packs, transparent props, y-sort placement, collision, zones, and previews
  • Engine Handoff: Godot scenes, editable TileMap layers, separated props, encounter grass, collision bodies, exits, and debug players
  • Local Cleanup: Chroma-key removal, frame extraction, alignment, transparent PNG/GIF export, prop-pack slicing, and QA metadata

Installation

# Clone the repository
git clone https://github.com/0x0funky/agent-sprite-forge.git
cd agent-sprite-forge

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
export OPENAI_API_KEY="your-api-key-here"
# or create .env file
echo "OPENAI_API_KEY=your-api-key-here" > .env

Requirements

  • Python 3.8+
  • PIL/Pillow for image processing
  • OpenAI API access (for image generation)
  • Optional: Godot 4.x or Unity for engine exports

Core Skills

1. Generate 2D Sprite Sheets ($generate2dsprite)

Use this skill when you need animated units, playable characters, monsters, props, spell bundles, or projectile/impact effects.

Basic sprite generation:

from agent_sprite_forge import SpriteGenerator

# Initialize generator
sprite_gen = SpriteGenerator()

# Generate character sprite sheet
result = sprite_gen.generate(
    prompt="samurai warrior walking animation, 4 directions, pixel art style",
    frames=4,
    directions=["down", "left", "right", "up"],
    output_dir="./output/sprites"
)

# Result includes:
# - sprite_sheet.png (full sheet)
# - frames/ (individual PNG frames)
# - animations/ (GIF animations per direction)
# - metadata.json (frame info, dimensions, cleanup settings)

Attack and spell animations:

# Generate spell cast animation
spell_result = sprite_gen.generate(
    prompt="fire mage casting fireball spell, side view, 6 frames",
    frames=6,
    animation_type="cast",
    include_projectile=True,
    output_dir="./output/spells/fireball"
)

# Generates:
# - cast_animation.gif
# - projectile_animation.gif
# - impact_animation.gif (if include_impact=True)
# - transparent PNG frames for each

Reference-guided variants:

# Generate sprite based on reference image
ref_result = sprite_gen.generate_from_reference(
    reference_image="./refs/crocodile.jpg",
    prompt="crocodile playing with a stone, 8 frame loop animation",
    frames=8,
    preserve_style=True,
    output_dir="./output/croc_animation"
)

2. Generate 2D Maps ($generate2dmap)

Use this skill for layered RPG maps with ground, props, collision, and engine-ready exports.

Basic layered map generation:

from agent_sprite_forge import MapGenerator

map_gen = MapGenerator()

# Generate layered map with prop extraction
map_result = map_gen.generate_layered_map(
    prompt="cyberpunk canal district, neon signs, bridges, water",
    style="hand-painted HD game map",
    include_props=True,
    output_dir="./output/maps/cyber_canal"
)

# Generates:
# - ground_base.png (clean ground layer)
# - dressed_reference.png (fully decorated reference)
# - prop_pack_3x3.png (extracted props grid)
# - props/ (individual transparent prop PNGs)
# - layered_preview.png (flattened preview)
# - metadata.json (layer info, prop positions)

Godot TileMap export:

# Generate Godot-ready map project
godot_map = map_gen.generate_godot_map(
    prompt="forest village with paths, houses, trees, pond",
    tileset_size=16,  # 16x16 tiles
    map_size=(40, 30),  # tiles
    include_collision=True,
    include_zones=True,  # encounter grass, exits
    output_dir="./output/godot_maps/forest_village"
)

# Generates:
# - tileset.png (autotile-ready)
# - prop_sheet_3x3.png
# - scenes/ForestVillage.tscn (editable scene)
# - resources/tileset.tres
# - resources/prop_atlas.tres
# - scripts/debug_player.gd
# - collision and area metadata

3. Post-Processing and Cleanup

Chroma-key background removal:

from agent_sprite_forge.processors import BackgroundRemover

remover = BackgroundRemover()

# Remove background and export transparent PNG
clean_sprite = remover.remove_background(
    input_path="./raw/sprite_sheet.png",
    chroma_color="#00FF00",  # green screen
    tolerance=30,
    output_path="./output/sprite_sheet_clean.png"
)

Frame extraction and GIF export:

from agent_sprite_forge.processors import FrameExtractor, GIFExporter

# Extract frames from sprite sheet
extractor = FrameExtractor()
frames = extractor.extract_frames(
    sprite_sheet="./output/sprite_sheet_clean.png",
    frame_width=64,
    frame_height=64,
    output_dir="./output/frames"
)

# Create animated GIF
gif_exporter = GIFExporter()
gif_exporter.create_animation(
    frames_dir="./output/frames",
    output_path="./output/animation.gif",
    frame_duration=100,  # milliseconds
    loop=0  # infinite loop
)

Prop pack slicing:

from agent_sprite_forge.processors import PropPackSlicer

slicer = PropPackSlicer()

# Slice 3x3 prop pack into individual props
props = slicer.slice_prop_pack(
    prop_pack="./output/maps/cyber_canal/prop_pack_3x3.png",
    grid_size=(3, 3),
    output_dir="./output/maps/cyber_canal/props",
    remove_background=True,
    chroma_color="#FFFFFF"
)

# Returns list of prop files with bounding box metadata

Configuration

Create a config.yaml file to set defaults:

# Image generation settings
image_generation:
  model: "dall-e-3"
  size: "1024x1024"
  quality: "hd"
  style: "natural"  # or "vivid"

# Sprite settings
sprites:
  default_frame_count: 4
  default_frame_size: [64, 64]
  background_color: "#00FF00"  # chroma key
  export_formats: ["png", "gif"]

# Map settings
maps:
  default_tile_size: 16
  prop_pack_grid: [3, 3]
  layer_order: ["ground", "props", "collision", "zones"]

# Processing
processing:
  chroma_tolerance: 30
  gif_frame_duration: 100
  png_compression: 9

# Engine exports
godot:
  version: "4.5"
  tilemap_layer_enabled: true
  include_debug_player: true

unity:
  version: "2022.3"
  sprite_pixels_per_unit: 16
  include_prefabs: true

Load config in Python:

from agent_sprite_forge import load_config

config = load_config("config.yaml")
sprite_gen = SpriteGenerator(config=config)

Common Patterns

Full Character Asset Pipeline

from agent_sprite_forge import SpriteGenerator, BackgroundRemover, FrameExtractor, GIFExporter

def create_character_assets(character_name, prompt_base):
    """Generate complete character asset set"""
    sprite_gen = SpriteGenerator()
    
    # 1. Generate directional walk animations
    walk_result = sprite_gen.generate(
        prompt=f"{prompt_base}, walking animation, 4 directions",
        frames=4,
        directions=["down", "left", "right", "up"],
        output_dir=f"./output/characters/{character_name}/walk"
    )
    
    # 2. Generate attack animation
    attack_result = sprite_gen.generate(
        prompt=f"{prompt_base}, sword attack, side view",
        frames=6,
        animation_type="attack",
        output_dir=f"./output/characters/{character_name}/attack"
    )
    
    # 3. Generate idle animation
    idle_result = sprite_gen.generate(
        prompt=f"{prompt_base}, idle breathing animation",
        frames=2,
        animation_type="idle",
        output_dir=f"./output/characters/{character_name}/idle"
    )
    
    return {
        "walk": walk_result,
        "attack": attack_result,
        "idle": idle_result
    }

# Usage
samurai_assets = create_character_assets(
    "samurai_warrior",
    "pixel art samurai in blue armor"
)

RPG Map with Godot Export

from agent_sprite_forge import MapGenerator

def create_rpg_zone(zone_name, description):
    """Generate RPG map zone with Godot export"""
    map_gen = MapGenerator()
    
    # Generate layered map
    map_result = map_gen.generate_godot_map(
        prompt=description,
        tileset_size=16,
        map_size=(40, 30),
        include_collision=True,
        include_zones=True,
        zone_types=["encounter_grass", "exit", "trigger"],
        output_dir=f"./output/maps/{zone_name}"
    )
    
    # Add custom metadata
    map_result.add_metadata({
        "zone_name": zone_name,
        "encounters": ["slime", "bat", "wolf"],
        "exits": [
            {"name": "north_exit", "to": "zone_02"},
            {"name": "south_exit", "to": "zone_00"}
        ]
    })
    
    return map_result

# Usage
forest_zone = create_rpg_zone(
    "forest_01",
    "dense forest with winding path, old shrine, mushroom clusters"
)

Spell Bundle Generation

def create_spell_bundle(spell_name, element, color_scheme):
    """Generate complete spell asset bundle"""
    sprite_gen = SpriteGenerator()
    
    spell_dir = f"./output/spells/{spell_name}"
    
    # Cast animation
    cast = sprite_gen.generate(
        prompt=f"mage casting {element} spell, {color_scheme} colors, side view",
        frames=6,
        output_dir=f"{spell_dir}/cast"
    )
    
    # Projectile
    projectile = sprite_gen.generate(
        prompt=f"{element} projectile, {color_scheme}, spinning animation",
        frames=4,
        animation_type="projectile",
        output_dir=f"{spell_dir}/projectile"
    )
    
    # Impact
    impact = sprite_gen.generate(
        prompt=f"{element} explosion impact, {color_scheme}, burst effect",
        frames=8,
        animation_type="impact",
        output_dir=f"{spell_dir}/impact"
    )
    
    return {
        "cast": cast,
        "projectile": projectile,
        "impact": impact
    }

# Usage
fireball = create_spell_bundle("fireball", "fire", "red and orange")
ice_lance = create_spell_bundle("ice_lance", "ice", "blue and white")

Unity Survivors-like Setup

from agent_sprite_forge import SpriteGenerator, UnityExporter

def setup_survivors_game(game_name):
    """Generate assets for survivors-like game"""
    sprite_gen = SpriteGenerator()
    unity_export = UnityExporter()
    
    base_dir = f"./output/unity_projects/{game_name}"
    
    # Hero sheets (4 directions)
    hero = sprite_gen.generate(
        prompt="anime hero character, walking animation, 4 directions",
        frames=4,
        directions=["down", "left", "right", "up"],
        output_dir=f"{base_dir}/Assets/Heroes/Hero01"
    )
    
    # Summon/minion sheets
    summons = []
    for i, summon_type in enumerate(["wolf", "eagle", "bear"]):
        summon = sprite_gen.generate(
            prompt=f"spirit {summon_type} companion, idle and attack",
            frames=6,
            output_dir=f"{base_dir}/Assets/Summons/{summon_type}"
        )
        summons.append(summon)
    
    # Enemy sheets
    enemies = []
    for enemy_type in ["slime", "skeleton", "demon"]:
        enemy = sprite_gen.generate(
            prompt=f"{enemy_type} enemy, walk and attack animation",
            frames=4,
            output_dir=f"{base_dir}/Assets/Enemies/{enemy_type}"
        )
        enemies.append(enemy)
    
    # Export Unity-ready structure
    unity_export.export_project(
        assets={
            "hero": hero,
            "summons": summons,
            "enemies": enemies
        },
        output_dir=base_dir,
        include_prefabs=True,
        include_animator_controllers=True
    )
    
    return base_dir

# Usage
game_path = setup_survivors_game("SummonSurvivors")

API Reference

SpriteGenerator

class SpriteGenerator:
    def generate(
        self,
        prompt: str,
        frames: int = 4,
        directions: list = None,
        animation_type: str = "walk",
        include_projectile: bool = False,
        include_impact: bool = False,
        output_dir: str = "./output"
    ) -> SpriteResult
    
    def generate_from_reference(
        self,
        reference_image: str,
        prompt: str,
        frames: int,
        preserve_style: bool = True,
        output_dir: str = "./output"
    ) -> SpriteResult

MapGenerator

class MapGenerator:
    def generate_layered_map(
        self,
        prompt: str,
        style: str = "hand-painted HD game map",
        include_props: bool = True,
        output_dir: str = "./output"
    ) -> MapResult
    
    def generate_godot_map(
        self,
        prompt: str,
        tileset_size: int = 16,
        map_size: tuple = (40, 30),
        include_collision: bool = True,
        include_zones: bool = True,
        zone_types: list = None,
        output_dir: str = "./output"
    ) -> GodotMapResult

BackgroundRemover

class BackgroundRemover:
    def remove_background(
        self,
        input_path: str,
        chroma_color: str = "#00FF00",
        tolerance: int = 30,
        output_path: str = None
    ) -> str

GIFExporter

class GIFExporter:
    def create_animation(
        self,
        frames_dir: str,
        output_path: str,
        frame_duration: int = 100,
        loop: int = 0
    ) -> str

Troubleshooting

Background removal leaves artifacts

# Increase tolerance for better removal
remover.remove_background(
    input_path="sprite.png",
    chroma_color="#00FF00",
    tolerance=50,  # Increase from default 30
    output_path="sprite_clean.png"
)

# Or use edge smoothing
remover.remove_background(
    input_path="sprite.png",
    chroma_color="#00FF00",
    tolerance=30,
    smooth_edges=True,
    edge_blur=2,
    output_path="sprite_clean.png"
)

Generated sprites don't align properly

# Use frame alignment processor
from agent_sprite_forge.processors import FrameAligner

aligner = FrameAligner()
aligned_frames = aligner.align_frames(
    frames_dir="./output/frames",
    alignment="center",  # or "bottom", "top"
    reference_point="auto",  # detect common ground point
    output_dir="./output/frames_aligned"
)

Godot export scene not loading

# Verify Godot version compatibility
godot_map = map_gen.generate_godot_map(
    prompt="village map",
    godot_version="4.5",  # Explicit version
    output_dir="./output/godot_maps/village"
)

# Check generated scene file
import os
assert os.path.exists("./output/godot_maps/village/scenes/Village.tscn")

Image generation quality issues

# Use HD quality and specific style guidance
sprite_gen = SpriteGenerator(
    image_model="dall-e-3",
    quality="hd",
    style="vivid"  # More saturated colors
)

result = sprite_gen.generate(
    prompt="detailed pixel art character, sharp edges, vibrant colors, 64x64 resolution",
    frames=4,
    output_dir="./output/sprites"
)

GIF animations too large

# Optimize GIF export
gif_exporter = GIFExporter()
gif_exporter.create_animation(
    frames_dir="./output/frames",
    output_path="./output/animation.gif",
    frame_duration=100,
    loop=0,
    optimize=True,  # Enable optimization
    colors=64,  # Reduce color palette (max 256)
    dither=False  # Disable dithering for smaller size
)

Prop extraction misses objects

# Adjust detection sensitivity
slicer = PropPackSlicer()
props = slicer.slice_prop_pack(
    prop_pack="./props.png",
    grid_size=(3, 3),
    output_dir="./output/props",
    min_object_size=10,  # Smaller minimum
    detection_threshold=128,  # Adjust alpha threshold
    remove_background=True
)

Environment Variables

# Required
export OPENAI_API_KEY="sk-..."

# Optional
export AGENT_SPRITE_FORGE_OUTPUT_DIR="./my_assets"
export AGENT_SPRITE_FORGE_CACHE_DIR="./.cache"
export AGENT_SPRITE_FORGE_LOG_LEVEL="INFO"  # DEBUG, INFO, WARNING, ERROR

# Godot export
export GODOT_EXECUTABLE_PATH="/usr/local/bin/godot"

# Unity export
export UNITY_EXECUTABLE_PATH="/Applications/Unity/Hub/Editor/2022.3.0f1/Unity.app/Contents/MacOS/Unity"

CLI Usage

# Generate sprite sheet
python -m agent_sprite_forge sprite \
  --prompt "knight walking animation" \
  --frames 4 \
  --directions down,left,right,up \
  --output ./output/knight

# Generate layered map
python -m agent_sprite_forge map \
  --prompt "desert oasis with palm trees" \
  --include-props \
  --output ./output/maps/oasis

# Generate Godot map
python -m agent_sprite_forge godot-map \
  --prompt "cave system with crystals" \
  --tileset-size 16 \
  --map-size 40,30 \
  --output ./output/godot/cave

# Remove background
python -m agent_sprite_forge remove-bg \
  --input ./raw/sprite.png \
  --output ./clean/sprite.png \
  --chroma "#00FF00" \
  --tolerance 30

# Create GIF
python -m agent_sprite_forge make-gif \
  --frames ./output/frames \
  --output ./animation.gif \
  --duration 100 \
  --loop 0

FAQ

What does the agent-sprite-forge-2d-game-assets skill do?

Generate 2D sprite sheets, animated GIFs, transparent PNGs, layered maps, and engine-ready prototypes for Godot/Unity using Codex-driven asset pipelines.

How do I install agent-sprite-forge-2d-game-assets?

Run npx skills add https://github.com/reason-machines/ai-agent-skills --skill agent-sprite-forge-2d-game-assets in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/agent-sprite-forge-2d-game-assets/ (personal) or .claude/skills/agent-sprite-forge-2d-game-assets/ (project).

Which AI agents can use agent-sprite-forge-2d-game-assets?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is agent-sprite-forge-2d-game-assets safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains agent-sprite-forge-2d-game-assets?

It is published by reason-machines in reason-machines/ai-agent-skills. It has 217 recorded installs.

More from reason-machines/ai-agent-skills

View all →

agency-agents-zh-skill

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents322 installs

agent-reach-internet-access

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents228 installs

agentmemory-persistent-memory

reason-machines/ai-agent-skills

Add persistent memory to AI coding agents using agentmemory - remembers context, preferences, and decisions across sessions

AI & Agents216 installs

awesome-agent-skills

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents214 installs

oh-my-openagent-orchestration

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents211 installs

tencentdb-agent-memory

reason-machines/ai-agent-skills

Local-first long-term memory for AI agents with symbolic short-term memory and layered long-term recall via a 4-tier progressive pipeline

AI & Agents204 installs

Popular media & creative skills

View category →

hyperframes

heygen-com/hyperframes

Read first to make, edit, animate or render any video with HyperFrames (HTML rendered to video), including launch films, explainers, music videos, charts, maps, vertical clips, title cards, captions and overlays. Holds the render contract, the work loop and what each kind of video needs.

Media & Creative1.2M installs

media-use

heygen-com/hyperframes

Get or make any media for a HyperFrames video - music, sound effects, images, icons, logos, voiceover, avatar and generated video, transcripts, cutouts, colour grades and looks - with HeyGen or free local tools, and mix placed audio. Also for footage that looks dark, flat or dull, or should be stylized or hidden.

Media & Creative1.1M installs

hyperframes-registry

heygen-com/hyperframes

About 450 ready-made HyperFrames blocks and components installed as editable source - transitions, overlays, titles, kinetic type, captions, charts, map overlays, 3D, camera moves. Use it before hand-building any named visual (CRT scanlines, glitch, film grain, a chart); also for hyperframes add and catalog.

Media & Creative1M installs

hyperframes-creative

heygen-com/hyperframes

Non-animation creative direction for HyperFrames videos. Use for design spec (frame.md / design.md) handling, palettes, typography, narration, beat planning, audio-reactive visuals, composition patterns, and brand / style decisions. For atomic motion patterns and scene blueprints, use hyperframes-animation.

Media & Creative916.5K installs

ai-video-generation

101-skills/superpowers

Generate AI videos with Google Veo, Seedance 2.0, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI. Models: Veo 3.1, Veo 3, Seedance 2.0, HappyHorse 1.0, Wan 2.5, Grok Imagine Video, OmniHuman, Fabric, HunyuanVideo. Capabilities: text-to-video, image-to-video, reference-to-video, video editing, lipsync, avatar animation, video upscaling, foley sound. Use for: social media videos, marketing content, explainer videos, product demos, AI avatars. Triggers: video generation, ai video, text to video, image to video, veo, animate image, video from image, ai animation, video generator, generate video, t2v, i2v, ai video maker, create video with ai, runway alternative, pika alternative, sora alternative, kling alternative, seedance, happyhorse

Media & Creative881.5K installs

ai-image-generation

101-skills/superpowers

Generate AI images with GPT-Image-2, FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI. Models: GPT-Image-2, FLUX Dev LoRA, FLUX.2 Klein LoRA, Gemini 3 Pro Image, Grok Imagine, Seedream 4.5, Reve, ImagineArt. Capabilities: text-to-image, image-to-image, inpainting, LoRA, image editing, upscaling, text rendering. Use for: AI art, product mockups, concept art, social media graphics, marketing visuals, illustrations. Triggers: flux, image generation, ai image, text to image, stable diffusion, generate image, ai art, midjourney alternative, dall-e alternative, text2img, t2i, image generator, ai picture, create image with ai, generative ai, ai illustration, grok image, gemini image, gpt image, openai image, chatgpt image

Media & Creative880.3K installs