WonderPhi Library for organizations — perpetual single-organization commercial license

Everything below is identical to the retail product. What's different: your license covers the entire buying organization (unlimited machines, VMs, containers, cloud instances) with no seat count, no annual renewal, no telemetry, no activation. Redistribution / SaaS / hosted-service exposure require a separate OEM arrangement.

Easy Setup — No Package Manager, No Account, No Cloud

Single Python file. Python 3.8+ standard library only — nothing to pip install. Run with:

python wpc-recipe.py

Sample plan files ship with the download and run out of the box. Pair with WonderPhi Compute to execute the generated plans in parallel.

WonderPhi Library — 25 ready-made batch recipes for the tools you already use

Twenty-five hand-written recipes that turn everyday command-line tools (ffmpeg, ImageMagick, 7-Zip, curl, git, PDF tools, GPU utilities) into parallel batch jobs. Point at a folder, get a plan file, hand it to WonderPhi Compute, watch every core work at once.

What it is, in plain English

A "recipe" is a small, opinionated Python function that takes a folder of inputs and writes a plain-text plan file. That plan says "run this exact command on this exact file", one line per input file. WonderPhi Compute reads the plan and runs every line in parallel. You end up doing an afternoon's manual work in the time it takes to make coffee.

How you use it — three steps

1. Run the recipe file. python wpc-recipe.py in the folder that contains your input files, or wherever you want the plan written.
2. Point at the target. The recipe asks (or reads from a text file) what the input is — a folder, a list of hostnames, a CSV, whatever the recipe expects. It writes a plan file: one shell command per line.
3. Hand the plan to Compute. Drop the generated plan.wpc.txt into the WonderPhi Compute inbox, and every command runs in parallel across your cores. Read the honest log when it's done.

Worked examples — three of the 25 recipes in this library

ffmpeg-transcode-batch — encode 200 videos to h264/mp4

Point at a folder of .mov/.avi/.mkv sources. Recipe emits a plan-line per file. Compute runs them across every core.

python wpc-recipe.py ffmpeg-transcode-batch --input ./raw --output ./mp4 --codec h264 --crf 22

# generates: plan-ffmpeg-2026-07-23.wpc.txt
# ffmpeg -i ./raw/clip01.mov -c:v libx264 -crf 22 ./mp4/clip01.mp4
# ffmpeg -i ./raw/clip02.mov -c:v libx264 -crf 22 ./mp4/clip02.mp4
# ... (one line per source file)

imagemagick-thumb-grid — make thumbnails of every image in a folder

One line per image. Runs concurrently. Comes back with 4,000 thumbnails in under a minute on a 40-core box.

python wpc-recipe.py imagemagick-thumb --input ./photos --output ./thumbs --size 320x320

# generates: plan-imagemagick-2026-07-23.wpc.txt
# magick photos/IMG_0001.jpg -resize 320x320 thumbs/IMG_0001.jpg
# magick photos/IMG_0002.jpg -resize 320x320 thumbs/IMG_0002.jpg
# ...

sha256-tree — verify integrity of a big backup

Emits a plan that hashes every file. WonderPhi Compute achieves ~14× speedup vs a serial script on typical mixed-file directories.

python wpc-recipe.py sha256-tree --input D:\backup-2026 --output backup-hashes.txt

# generates: plan-sha256-2026-07-23.wpc.txt
# certutil -hashfile "D:\backup-2026\file001.dat" SHA256 >> backup-hashes.txt
# certutil -hashfile "D:\backup-2026\file002.dat" SHA256 >> backup-hashes.txt
# ...

Under the hood — the honest technical picture

The recipe file is one .py file, ~400 lines. Each recipe is a small function that takes a target folder plus a few flags and emits a plan file to stdout or to disk. No hidden state, no plugin system, no config file — if a recipe misbehaves, you can literally read the 15 lines that produced the output.

• Standard library only. No dependencies, no pip install, no supply-chain risk.
• Plans are plain text. Every emitted plan is human-readable. You can edit one by hand before running it if you want to tweak.
• Recipes are opinionated but small. Each does one job well. If a recipe doesn't fit your exact case, copy it and modify — every recipe is under 30 lines of code.
• No mutation of your inputs. Recipes only read the input folder. Every line in the emitted plan writes to a separate output folder that you specify.
• Deterministic output. Same inputs plus same flags = same plan file. Diff-friendly for source-control workflows.

How it compares to the manual way

What you need

• Python 3.8+ to run the recipe file. That's it — standard library only.
• The underlying tool the recipe wraps (ffmpeg, ImageMagick, etc.) — recipes call them, they don't ship them.
• WonderPhi Compute to run the generated plan in parallel. (Sold separately, or bundle everything with the Complete Bundle.)
• Windows, macOS, or Linux. Individual recipes may need a specific OS — e.g. an ImageMagick recipe needs magick in your PATH.

What's in the download

• wpc-recipe.py — the recipe generator (Python 3.8+, standard library only)
• README.md — per-recipe reference for humans
• INDEX.md — alphabetical list of every recipe with a one-line description
• samples/ — ready-to-run demo input files for every recipe
• LICENSE.txt — perpetual single-user license, personalized with your name and order number at delivery

Common questions

License & support — Industrial variant

Perpetual, non-exclusive, non-transferable Industrial license. Covers all machines, VMs, containers, and cloud instances operated by a single organization (the legal entity named on the purchase receipt plus wholly-owned subsidiaries). No seat count, no machine limit, no renewals, no activation, no telemetry.

Employees, contractors, and consultants under the organization's direction may operate the software solely for the licensed organization's benefit. Rights end when their engagement ends.

Not permitted without a separate arrangement: redistributing the software; hosting it publicly; embedding it in a product/SaaS/PaaS you sell to third parties; sublicensing.

Support: email chris@cri-one.com.

Every download's LICENSE.txt is personalized at checkout with your organization name, order number, and issue date. USA sales.