why-cli 1.3.3

Track why you installed every tool on your machine.

whydatApp (why?)

Ever install something, then months later wonder why it's on your machine?

why? catches every package manager install (brew, npm, pip, cargo, apt, …) right as it happens via a lightweight shell hook, then prompts you to jot down what the tool is for, which project needed it, whether it's worth keeping, and what to do with it later. Everything stays in a local SQLite database — nothing is sent anywhere. Browse, search, filter, and export your install history through a built-in web UI, or right from the terminal.

What you can answer with why?:

• What do I have installed? — Full inventory across all package managers in one place
• Why did I install this? — Context captured at install time, while you remember
• Which project needed it? — Auto-inferred from your working directory
• Is this still useful? — Mark tools as experimental, for removal, or permanent docs
• What's taking up space? — See resolved install paths and identify candidates to uninstall
• Did I ever install X? — Search by name, command, or description — find it instantly
• What should I clean up? — Review queue surfaces incomplete entries and stale experimental installs
• How do I set up this project elsewhere? — Export your setup dependencies to Markdown or JSON
• Which managers do I use most? — Dashboard shows install trends by manager, project, and month

Install

From PyPI (recommended):

uv tool install 'why-cli[web]'   # or: pipx install 'why-cli[web]'
why init                          # interactive setup; edits your shell rc

From source (for development or pre-release):

git clone https://github.com/Nostoi/whydatapp.git
cd whydatapp
uv tool install --editable '.[web]'
why init

Restart your shell, then try brew install ripgrep (or any tracked manager).

Full install instructions, including building from a wheel: Install guide.

Quick reference

Command	What it does
why init	Interactive first-run setup
why	Show help and the available subcommands
why log -- <cmd>	Manually log an install
why review	Drain the skipped/incomplete review queue
why list	Print installs as a table
why export	Export to Markdown or JSON
why serve	Open the local web UI at 127.0.0.1:7873
why uninstall	Remove the hook (and optionally the data)

Detailed usage with examples: Usage guide.

Documentation

• Install — Requirements, install paths (PyPI / source / wheel), what why init does, uninstall.
• Usage — Every CLI subcommand with examples.
• Web UI — Walkthrough of the local web interface.
• Configuration — ~/.why/*.toml files, env vars, ignore rules.
• Troubleshooting — Hook not firing, prompt missed, address-in-use, restoring data, filing bugs.
• Development — Clone, set up, run tests, build the wheel, project layout, contribute.
• Design spec — Architecture, data model, decisions, post-MVP roadmap.

Privacy

• All data lives in ~/.why/data.db. No network calls.
• The web UI binds to 127.0.0.1 only.
• All static assets vendored locally — no CDN, no Google Fonts, no analytics.
• The shell hook ignores any install triggered by another tracked installer (no false positives from brew resolving deps).

License

MIT