65-wonderphi-library

Valuation

Generous asset valuation: $10,000,000,000. The listed price is the platform maximum; acquisition at valuation is handled by direct enquiry.

WonderPhi Library v1.0.0

WonderPhi Library v1.0.0

The Recipe Library for WonderPhi Compute — pre-optimized plans for the

programs you already use.

You have 500 videos to transcode. You don't sit down and write a plan.

You run one line:

python wpc-recipe.py ffmpeg-h264 .\videos

…and out comes videos.plan.wpc.txt — one tuned command per file, ready to

feed straight to WonderPhi Compute, which runs them across every core you own.

What this is (in plain English)

WonderPhi Compute runs a plan — a plain-text

list of commands — across all your CPU cores at once. It's fast and honest,

but someone still has to write the plan.

WonderPhi Library writes the plan for you. It ships a curated set of

recipes for the most popular programs (ffmpeg, ImageMagick, PDF tools,

7-Zip, hashing, curl, git, and the obvious NVIDIA GPU jobs). Point a recipe at

a folder, and it emits a WonderPhi Compute plan with the good flags already

filled in — the ones you'd have to look up on Stack Overflow otherwise.

• Library writes the plan. WonderPhi Compute runs the plan.
• Nothing here executes your files — it only reads a folder and writes a .txt.
• Zero dependencies. Python standard library only. No network. No telemetry.
• It never modifies wpc.exe. The 10 Commandments stay intact — this is pure

convenience layered on top of the runtime.

Three steps

1.  python wpc-recipe.py --list                 # see every recipe
2.  python wpc-recipe.py <recipe> <folder>      # generate the plan
3.  wpc.exe <that-plan>.plan.wpc.txt            # WonderPhi Compute runs it

Or, if you have WonderPhi Compute's always-on watcher installed, add --inbox

and the plan is dropped straight into wpc-inbox\ to run itself:

python wpc-recipe.py img-webp .\photos --inbox

Examples

# Re-encode a folder of videos to H.264 (CPU)
python wpc-recipe.py ffmpeg-h264 .\raw --outdir .\mp4

# Same, but on the GPU via NVENC (much faster; needs an NVIDIA card)
python wpc-recipe.py ffmpeg-nvenc .\raw --outdir .\mp4

# Convert every image in a folder to WebP
python wpc-recipe.py img-webp .\photos --outdir .\web

# Extract text from a pile of PDFs
python wpc-recipe.py pdf-text .\reports --outdir .\text

# SHA-256 every file, with a combined manifest at the end
python wpc-recipe.py sha256 .\evidence --outdir .\hashes

# Health-check a list of URLs (one per line in urls.txt)
python wpc-recipe.py curl-check .\urls.txt

# Transcribe a folder of recordings with Whisper on the GPU
python wpc-recipe.py whisper .\recordings --outdir .\transcripts

# Run every test suite under .\projects in parallel
python wpc-recipe.py pytest-dirs .\projects

GPU recipes (the obvious ones)

Marked [GPU] in --list. They emit NVIDIA-accelerated command lines:

--gpu auto (the default) probes nvidia-smi. If you have no GPU, the tool

warns and points you at the CPU equivalent (e.g. ffmpeg-nvenc → ffmpeg-h264).

When WonderPhi Compute's watcher runs a GPU plan, it also exposes

WPC_GPGPU, WPC_GPU_VRAM_FREE_MB, and WPC_GPU_UTIL_PCT to every task.

How a recipe reads its input

Every recipe uses one of three input models (shown in --list):

Options

What's in the box

65-wonderphi-library/
├── wpc-recipe.py     ← the generator (Python, stdlib only)
├── README.md         ← this file
├── INDEX.md          ← full recipe table
├── LICENSE
└── samples/
    ├── example-urls.txt
    └── curl-check.example.plan.wpc.txt

Requires Python 3.8+ to generate a plan, and WonderPhi Compute to run it.

The individual programs a recipe drives (ffmpeg, magick, 7z, …) must be on your

PATH — each recipe's Needs on PATH: header tells you which one.

Christopher Gabriel Brown · Inventor · Author · Visionary · cri-one.com

WonderPhi Library — Recipe Index

WonderPhi Library — Recipe Index

25 recipes. Generated from the registry in wpc-recipe.py.

Models: files = scan a folder for matching files · dirs = one job per subfolder · lines = one job per line of a list file.

Generate any of these with:

python wpc-recipe.py <recipe> <input> [--gpu auto|on|off] [--outdir DIR]

This archive contains 2 documents. The complete folder ships as the product.