Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/medy-gribkov/arcananpx agentmods add skills/medy-gribkov/arcana/asset-optimizationWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/medy-gribkov/arcana/asset-optimization)<a href="https://agentmods.dev/skills/medy-gribkov/arcana/asset-optimization"><img src="https://agentmods.dev/badge/skills/medy-gribkov/arcana/asset-optimization/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/medy-gribkov/arcana/asset-optimization"><img src="https://agentmods.dev/badge/skills/medy-gribkov/arcana/asset-optimization.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00029 | $0.02460 |
| Opus 5 | $0.00015 | $0.01230 |
| Sonnet 5 | $0.00006 | $0.00492 |
| Haiku 4.5 | $0.00003 | $0.00246 |
Grade A, and why
asset-optimization scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
subprocess.run(cmd, check=True, capture_output=True) How it starts
The opening of the file, as written. The whole thing — 343 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Asset Optimization
Workflow Overview
Asset optimization follows this 4-step pattern:
- Audit - Identify largest assets, measure load times
- Compress - Apply platform-specific compression
- Convert - Transform to efficient runtime formats
- Validate - Verify quality and measure savings
Texture Optimization Workflow
Step 1: Audit Current Usage
# Find all textures and sort by size
find ./Assets/Textures -type f \( -name "*.png" -o -name "*.tga" \) -exec ls -lh {} \; | sort -k5 -hr | head -20
# Output total size
du -sh ./Assets/Textures
Step 2: Apply Compression
BAD - Uncompressed RGBA32:
Character_Diffuse.png: 2048x2048, RGBA32
Size: 16 MB
Memory: 16 MB at runtime
GOOD - Platform-optimized BC7:
Character_Diffuse.dds: 2048x2048, BC7
Size: 2 MB
Memory: 2 MB at runtime (8x savings)
Step 3: Batch Convert with texconv
# Windows - Convert all PNG to BC7 DDS with mipmaps
for file in ./source/*.png; do
texconv -f BC7 -m 0 -o ./optimized/ "$file"
done
# Mobile - Convert to ASTC 6x6
for file in ./source/*.png; do
astcenc -cl "$file" ./optimized/$(basename "$file" .png).astc 6x6 -medium
done
Step 4: Validate Quality
# Compare file sizes
du -sh ./source ./optimized
# Visual diff (requires ImageMagick)
compare -metric PSNR source.png optimized.png diff.png
WebP/AVIF Conversion for Web
WebP Conversion
# Single file - 80% quality, lossless alpha
cwebp -q 80 input.png -o output.webp
# Batch convert entire directory
find ./images -name "*.png" -exec bash -c 'cwebp -q 80 "$0" -o "${0%.png}.webp"' {} \;
# With fallback generation
for img in ./images/*.png; do
cwebp -q 80 "$img" -o "${img%.png}.webp"
# Keep original as fallback
done
Before/After Example:
hero-banner.png: 1.2 MB (PNG, lossless)
hero-banner.webp: 180 KB (WebP, 85% savings)
AVIF Conversion (Better Compression)
# Install avif encoder
npm install -g @squoosh/cli
# Convert with quality 60 (good balance)
squoosh-cli --avif '{"cqLevel":60}' input.png
# Batch process
find ./images -name "*.png" | xargs squoosh-cli --avif '{"cqLevel":60}'
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 343 lines · 29 tokens per session scan A 8745f5c53db7
asset-optimization is a skill published in the GitHub repository medy-gribkov/arcana (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 29 tokens to every session and 2,460 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
pixel-art
Design pixel art sprites, tilesets, and animations — game assets, retro icons, and character sheets.
ar-vr-xr
AR/VR/XR development with Unity XR, WebXR, ARKit, ARCore, Meta Quest SDK, and spatial computing. Use when building augmented reality, virtual reality, mixed reality applications, or spatial experiences.
hue
Meta-skill that generates new design language skills. Works on Claude Code and Codex. Use when the user says 'create a design skill', 'generate design language', 'new design system skill', 'design skill inspired by X', 'design skill from this screenshot', '/hue', or 'use hue'. Also triggers for 'remix my design skill'…
unity-ai-behavior
Use when implementing AI and NPC behavior — state machines, behavior trees, GOAP, NavMesh navigation, decision-making patterns, and when to use each approach.
unity-performance
Use when analyzing or optimizing Unity performance — CPU/GPU/RAM profiling, object pooling, batching strategies, physics optimization, memory management, and platform-specific budgets.
unity-testing
Use when writing or structuring tests — EditMode vs PlayMode test decisions, test architecture, what to test in Unity, mocking strategies, and CI integration for Unity tests.