ASCII Art

A decision-routing framework for text-based ASCII art. Each request type maps to a different tool — this skill picks the right one.

Tool Decision Tree

What do you need?
├── Large text banner from a string
│   ├── Need 500+ font choices, no install → asciified API (remote)
│   ├── Local Python available → pyfiglet (local, 571 fonts)
│   └── Want color/ANSI effects → TOIlet (Linux/WSL only)
│
├── Speech bubble or fun ASCII character
│   └── cowsay (50+ characters)
│
├── Decorative border around text block
│   └── boxes (70+ border styles)
│
├── Convert an image to ASCII
│   ├── Supports color, braille, URLs → ascii-image-converter
│   └── Lightweight JPEG only → jp2a
│
├── Browse pre-made art by subject
│   └── ascii.co.uk (searchable collection)
│
├── QR code, weather art, GitHub Octocat
│   └── Fun APIs: qrenco.de, wttr.in, github.com/octocat
│
└── Custom / conceptual art (not covered above)
    └── LLM-generated using Unicode palette (see below)

Tool Reference

pyfiglet (local — Python)

pip install pyfiglet
python -c "import pyfiglet; print(pyfiglet.figlet_format('Hello', font='slant'))"
# List available fonts:
python -c "import pyfiglet; print('\n'.join(pyfiglet.FigletFont.getFonts()[:20]))"

571 built-in fonts. Good defaults: slant, banner3, big, doom, isometric1, roman.

asciified API (remote — no install)

curl "https://asciified.thelicato.io/api/v2/ascii?text=Hello&font=slant"

250+ FIGlet fonts, free REST endpoint, works anywhere curl is available.

TOIlet (local — Linux/WSL only)

# Install: sudo apt install toilet
toilet -f mono9 -F border "Hello"
toilet -f future --gay "Hello"    # rainbow ANSI
toilet --list                     # show available fonts

Windows note: Requires WSL. Not available in native PowerShell/CMD.

cowsay (local)

# Install: brew install cowsay / sudo apt install cowsay
cowsay "Hello, World"
cowsay -f dragon "Rawr"
cowsay -l               # list 50+ characters
cowsay -e "@@" -T "U"   # custom eyes and tongue

Characters include: cow, dragon, tux, ghostbusters, stegosaurus, vader, and 45+ more.

boxes (local)

# Install: brew install boxes / sudo apt install boxes
echo "Hello" | boxes -d stone
echo "Hello" | boxes -d dog -a c    # centered
boxes -l                             # list 70+ designs

Good designs: stone, simple, parchment, dog, scroll, columns, peek.

ascii-image-converter (local)

# Install: go install github.com/TheZoraiz/ascii-image-converter@latest
ascii-image-converter image.jpg
ascii-image-converter image.jpg --color        # ANSI color
ascii-image-converter image.jpg --braille      # braille charset
ascii-image-converter "https://url/image.jpg"  # from URL
ascii-image-converter image.jpg -W 80          # 80-char width

jp2a (local — JPEG only)

# Install: brew install jp2a / sudo apt install jp2a
jp2a image.jpg
jp2a --width=60 image.jpg

Fun APIs (remote)

# QR code
curl "qrenco.de/https://example.com"

# Weather art
curl "wttr.in/London"
curl "wttr.in/London?format=3"      # compact

# GitHub Octocat with quote
curl "https://api.github.com/octocat"

LLM-Generated Custom Art

When no tool fits (conceptual art, diagrams, logos from text), generate directly using this character palette:

Block characters: █ ▓ ▒ ░
Box drawing: ─ │ ┌ ┐ └ ┘ ├ ┤ ┬ ┴ ┼ ═ ║ ╔ ╗ ╚ ╝
Geometric: ▲ ▼ ◆ ● ○ ◉ ★ ☆
Braille dots: ⠁ ⠂ ⠃ ⠄ ⠅ ⠆ ⠇ ⠈ ⠉ ⠊
Lines/arrows: ← → ↑ ↓ ↔ ↕ ⟵ ⟶ ⇒ ⟹
Math/misc: ∞ ∑ ∏ √ ∂ ∫ ≈ ≠ ≤ ≥

Best practices for LLM-generated art:

• Use monospace mental model: each character is equal width
• Plan on paper/mentally before drawing
• Start with the outline, fill interior second
• Keep width ≤ 80 chars for readability in most terminals
• Use pre or code blocks when displaying in Markdown

Platform Notes

Tool	Windows	macOS	Linux
pyfiglet	✓ (pip)	✓	✓
asciified API	✓ (curl)	✓	✓
TOIlet	WSL only	brew	apt
cowsay	WSL/chocolatey	brew	apt
boxes	WSL only	brew	apt
ascii-image-converter	✓ (go)	✓	✓
jp2a	WSL/pre-built	brew	apt
Fun APIs	✓ (curl/pwsh)	✓	✓