python-scripts

name	python-scripts
description	Run Python scripts in an isolated virtual environment. Use when you need to execute Python code for image manipulation, data processing, or any scripting task. Use when this capability is needed.
metadata	{"author":"fnsk4r17s"}

Python Scripts Skill

Run Python scripts in an isolated virtual environment that is created and cleaned up automatically.

When to Use

• Image manipulation (resize, crop, transparency, format conversion)
• Data processing and file manipulation
• Quick scripting tasks that need Python libraries
• Any task requiring PIL/Pillow, numpy, or other common Python packages

Execution Pattern

Always use this pattern to run Python scripts:

# Create venv, install dependencies, run script, cleanup
python3 -m venv /apps/chakravarti-cli/.tmp_venv && \
source /apps/chakravarti-cli/.tmp_venv/bin/activate && \
pip install -q <packages> && \
python3 << 'EOF'
# Your Python code here
EOF
rm -rf /apps/chakravarti-cli/.tmp_venv

Example: Image Manipulation

python3 -m venv /apps/chakravarti-cli/.tmp_venv && \
source /apps/chakravarti-cli/.tmp_venv/bin/activate && \
pip install -q pillow numpy && \
python3 << 'EOF'
from PIL import Image
import numpy as np

# Load image
img = Image.open('/path/to/image.png').convert('RGBA')

# Manipulate...
# img.resize((128, 128), Image.Resampling.LANCZOS)

# Save
img.save('/path/to/output.png')
print('Done!')
EOF
rm -rf /apps/chakravarti-cli/.tmp_venv

Example: Make Corners Transparent

python3 -m venv /apps/chakravarti-cli/.tmp_venv && \
source /apps/chakravarti-cli/.tmp_venv/bin/activate && \
pip install -q pillow numpy && \
python3 << 'EOF'
from PIL import Image
import numpy as np

img = Image.open('/path/to/logo.png').convert('RGBA')
data = np.array(img)
height, width = data.shape[:2]

# Corner radius as percentage of image size
corner_radius = int(min(width, height) * 0.12)

for y in range(height):
    for x in range(width):
        in_corner = False
        
        # Top-left
        if x < corner_radius and y < corner_radius:
            if ((corner_radius - x)**2 + (corner_radius - y)**2)**0.5 > corner_radius:
                in_corner = True
        # Top-right
        if x >= width - corner_radius and y < corner_radius:
            if ((x - (width - corner_radius - 1))**2 + (corner_radius - y)**2)**0.5 > corner_radius:
                in_corner = True
        # Bottom-left
        if x < corner_radius and y >= height - corner_radius:
            if ((corner_radius - x)**2 + (y - (height - corner_radius - 1))**2)**0.5 > corner_radius:
                in_corner = True
        # Bottom-right
        if x >= width - corner_radius and y >= height - corner_radius:
            if ((x - (width - corner_radius - 1))**2 + (y - (height - corner_radius - 1))**2)**0.5 > corner_radius:
                in_corner = True
        
        if in_corner:
            data[y, x, 3] = 0  # Set alpha to 0

Image.fromarray(data).save('/path/to/output.png')
print('Corners made transparent!')
EOF
rm -rf /apps/chakravarti-cli/.tmp_venv

Example: Resize for App Icons

python3 -m venv /apps/chakravarti-cli/.tmp_venv && \
source /apps/chakravarti-cli/.tmp_venv/bin/activate && \
pip install -q pillow && \
python3 << 'EOF'
from PIL import Image

logo = Image.open('/apps/chakravarti-cli/logo.png').convert('RGBA')

sizes = [
    (512, 'icon.png'),
    (256, '[e-mail]'),
    (128, '128x128.png'),
    (32, '32x32.png'),
]

icons_dir = '/apps/chakravarti-cli/crates/ckrv-tauri/icons/'

for size, filename in sizes:
    resized = logo.resize((size, size), Image.Resampling.LANCZOS)
    resized.save(icons_dir + filename)
    print(f'Created {filename} ({size}x{size})')
EOF
rm -rf /apps/chakravarti-cli/.tmp_venv

Common Packages

• pillow - Image manipulation
• numpy - Numerical operations, array manipulation
• requests - HTTP requests
• pyyaml - YAML parsing
• toml - TOML parsing

Important Notes

1. Always cleanup: Include rm -rf /apps/chakravarti-cli/.tmp_venv at the end
2. Use quiet install: Use pip install -q to reduce output noise
3. Heredoc for scripts: Use << 'EOF' syntax for multi-line Python scripts
4. RGBA for transparency: Always use .convert('RGBA') when working with transparency

---

Source: fnsk4r17s/chakravarti-cli — distributed by TomeVault.