agent-sprite-forge
```markdown
What this skill does
```markdown
---
name: agent-sprite-forge
description: Generate game-ready 2D sprite sheets, transparent PNG frames, and animated GIFs from natural-language prompts using the Codex agent skill $generate2dsprite.
triggers:
- generate a 2D sprite
- create a sprite sheet
- make a pixel art animation
- use generate2dsprite to create
- create an animated GIF from a prompt
- generate a sprite bundle with cast projectile impact
- make a four-direction walk sprite sheet
- create game assets from a text prompt
---
# Agent Sprite Forge
> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.
Agent Sprite Forge is a Codex agent skill (`$generate2dsprite`) that turns natural-language prompts into game-ready 2D pixel-art assets. The agent plans the asset, calls Codex's built-in image generation, then runs a deterministic local Python post-processor to clean up backgrounds, split frames, align cells, and export transparent PNGs and animated GIFs.
---
## How It Works
1. **Agent plans** the sheet layout, frame count, view angle, and action type.
2. **Codex generates** the raw sprite sheet image (built-in image generation — no external API needed).
3. **Local post-processor** (`generate2dsprite.py`) performs:
- Magenta (`#FF00FF`) chroma-key background removal
- Frame splitting based on grid layout
- Bounding-box extraction and alignment
- Rescaling
- Transparent PNG and animated GIF export
---
## Installation
### macOS / Linux
```bash
git clone https://github.com/0x0funky/agent-sprite-forge.git
cd ./agent-sprite-forge
python3 -m pip install -r ./requirements.txt
mkdir -p ~/.codex/skills
cp -R ./skills/generate2dsprite ~/.codex/skills/generate2dsprite
```
### Windows PowerShell
```powershell
git clone https://github.com/0x0funky/agent-sprite-forge.git
cd .\agent-sprite-forge
python -m pip install -r .\requirements.txt
New-Item -ItemType Directory -Force -Path "$env:USERPROFILE\.codex\skills" | Out-Null
Copy-Item -Recurse -Force `
".\skills\generate2dsprite" `
"$env:USERPROFILE\.codex\skills\generate2dsprite"
```
Start a **new Codex session** after installation so the skill is loaded cleanly.
### Python Dependencies
```text
Pillow
numpy
```
Install manually if needed:
```bash
pip install Pillow numpy
```
---
## Repository Layout
```text
agent-sprite-forge/
requirements.txt
skills/
generate2dsprite/
SKILL.md
agents/
openai.yaml # Codex agent config
references/
modes.md # Asset type / mode reference
prompt-rules.md # Prompt authoring rules
scripts/
generate2dsprite.py # Local post-processor
```
---
## Invoking the Skill
In a Codex session, use `$generate2dsprite` in your prompt. The agent reads `SKILL.md`, plans the asset, generates the sheet, then calls the local script.
### Basic invocation pattern
```text
Use $generate2dsprite to create <description>.
```
---
## Prompt Examples by Asset Type
### Idle / Creature
```text
Use $generate2dsprite to create a 3x3 idle for an ultimate earth titan.
```
```text
Use $generate2dsprite to create a golden divine boar 2x2 idle animation.
```
### Character Attack
```text
Use $generate2dsprite to create a side-view lightning knight attack animation.
```
```text
Use $generate2dsprite to create Omegamon attack and right-move animation assets.
```
### Spell Bundle (Cast + Projectile + Impact)
```text
Use $generate2dsprite to create a fire mage cast animation with projectile and impact.
```
```text
Use $generate2dsprite to create a wizard spell bundle with cast, projectile, and impact sprites.
```
```text
Use $generate2dsprite to create a fireball projectile loop and a matching explosion impact.
```
### Four-Direction Walk Sheet (top-down RPG)
```text
Use $generate2dsprite to create a top-down 4x4 player_sheet for a wandering young samurai
with a red scarf and short katana. Make a four-direction walk sprite sheet with 4 frames
per direction. Row order: down, left, right, up. Same character, same outfit, same
proportions, same pixel scale in every frame. Solid #FF00FF background. Each frame must
fit fully inside its cell, with clear margin on all sides. Retro JRPG pixel-art style.
```
### FX / Summon Effect
```text
Use $generate2dsprite to create a side-view summon entrance effect for a thunder wolf spirit.
```
### Reference Image → Sprite
Attach or reference an image in your Codex session:
```text
Use $generate2dsprite to create this male character teaching animation.
```
```text
Use $generate2dsprite — make an animation of this crocodile playing with the stone in its hand.
```
### Full Playable Game (One-Shot)
```text
Use $generate2dsprite to create a 2D side-scrolling game similar to Mega Man. It should
include attack mechanics, map elements, and all essential features. Design everything and
create all assets using this skill. It must be an actually playable game with a cyberpunk
story setting.
```
---
## What the Post-Processor Produces
For a typical sprite sheet, the output directory contains:
| File | Description |
|---|---|
| `raw-sheet.png` | Original image from Codex generation |
| `raw-sheet-clean.png` | After chroma-key removal |
| `sheet-transparent.png` | Full sheet with transparency |
| `frame-00.png`, `frame-01.png`, … | Individual extracted frames |
| `animation.gif` | Assembled animated GIF |
| `prompt-used.txt` | The exact prompt sent to generation |
| `pipeline-meta.json` | Grid layout, frame count, frame size metadata |
For **player walk sheets**, additional per-direction outputs are generated:
| File | Description |
|---|---|
| `strip-down.png` | Horizontal strip for down direction |
| `strip-left.png` | Horizontal strip for left direction |
| `strip-right.png` | Horizontal strip for right direction |
| `strip-up.png` | Horizontal strip for up direction |
| `down.gif` | Animated GIF for down walk |
| `left.gif` | Animated GIF for left walk |
| `right.gif` | Animated GIF for right walk |
| `up.gif` | Animated GIF for up walk |
---
## Running the Post-Processor Directly
If you have a raw sheet image and want to run post-processing manually:
```python
# Example: split a 3x3 sheet and export transparent frames + GIF
from PIL import Image
import numpy as np
def remove_magenta(img: Image.Image, tolerance: int = 30) -> Image.Image:
"""Replace #FF00FF background with transparency."""
rgba = img.convert("RGBA")
data = np.array(rgba, dtype=np.int32)
r, g, b, a = data[..., 0], data[..., 1], data[..., 2], data[..., 3]
mask = (
(r > 255 - tolerance) &
(g < tolerance) &
(b > 255 - tolerance)
)
data[..., 3] = np.where(mask, 0, a)
return Image.fromarray(data.astype(np.uint8), "RGBA")
def split_sheet(sheet_path: str, cols: int, rows: int, output_dir: str):
"""Split a sprite sheet into individual transparent frame PNGs and a GIF."""
import os
os.makedirs(output_dir, exist_ok=True)
sheet = Image.open(sheet_path).convert("RGBA")
total_w, total_h = sheet.size
frame_w = total_w // cols
frame_h = total_h // rows
frames = []
for row in range(rows):
for col in range(cols):
left = col * frame_w
top = row * frame_h
cell = sheet.crop((left, top, left + frame_w, top + frame_h))
clean = remove_magenta(cell)
idx = row * cols + col
out_path = os.path.join(output_dir, f"frame-{idx:02d}.png")
clean.save(out_path)
frames.append(clean)
print(f"Saved {out_path}")
# Export animated GIF
gif_path = os.path.join(output_dir, "animation.gif")
if frames:
frames[0].save(
gif_path,
save_all=True,
append_images=frames[1:],
loop=0,
duration=100, # ms per frame
disposal=2,
)
print(f"Saved {gif_path}")
# Usage
split_sheet("raw-sheet.png", cols=3, rows=3, outpRelated in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.