Learn-skills.dev image-upscale
Upscales an image using AI super-resolution to increase resolution with detail generation. Use when you need to enlarge images, improve low-resolution photos, or prepare images for large-format display.
install
source · Clone the upstream repo
git clone https://github.com/NeverSight/learn-skills.dev
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/NeverSight/learn-skills.dev "$T" && mkdir -p ~/.claude/skills && cp -r "$T/data/skills-md/agntswrm/agent-media/image-upscale" ~/.claude/skills/neversight-learn-skills-dev-image-upscale && rm -rf "$T"
manifest:
data/skills-md/agntswrm/agent-media/image-upscale/SKILL.mdsource content
Image Upscale
Upscales an image using AI super-resolution models, increasing resolution while generating realistic detail.
Command
agent-media image upscale --in <path> [options]
Inputs
| Option | Required | Description |
|---|---|---|
| Yes | Input file path or URL |
| No | Scale factor (2 or 4, default: 2) |
| No | Output path, filename or directory (default: ./) |
| No | Provider to use (local, fal, replicate) |
| No | Model to use (overrides provider default) |
Output
Returns a JSON object with the upscaled image path:
{ "ok": true, "media_type": "image", "action": "upscale", "provider": "local", "output_path": "upscaled_123_abc.png", "mime": "image/png", "bytes": 456789 }
Examples
Upscale an image (default 2x):
agent-media image upscale --in photo.jpg
Upscale 4x with specific provider:
agent-media image upscale --in photo.jpg --scale 4 --provider fal
Upscale with custom output:
agent-media image upscale --in photo.jpg --out ./upscaled
Providers
local
Runs locally on CPU using Transformers.js, no API key required.
- Uses
model (~1.3MB)Xenova/swin2SR-compressed-sr-x4-48 - Always outputs 4x upscale regardless of
(model architecture limitation)--scale - Models downloaded on first use
- You may see a
error — ignore it, the output is correct ifmutex lock failed"ok": true
agent-media image upscale --in photo.jpg --provider local
fal
- Requires
FAL_API_KEY - Uses
(Real-ESRGAN) modelfal-ai/esrgan - Supports 2x and 4x scale
replicate
- Requires
REPLICATE_API_TOKEN - Uses
modelnightmareai/real-esrgan - Supports 2-10x scale