picoexit

Pico CSS was archived by its maintainer on 2026-10-04 — v2.1.1 is the final release, forever. picoexit audits your project's real Pico footprint (which selectors, which components, which version), flags the selectors most likely to rot as browsers move on, and gives you a concrete exit plan: stay pinned, migrate, or follow a community fork. For migrations it ships a codemod that rewrites the most common Pico classes to Tailwind v4, Bulma v1, or plain CSS + Open Props — always as a diff preview first, never touching your files without --apply.

Install

npm is not published yet — install from source (Node.js 20+):

git clone https://github.com/techdevfun/picoexit.git
cd picoexit
npm install        # dev dependencies only (build/test); zero runtime dependencies
npm run build
node dist/src/cli.js audit ./my-site   # scan a project
node dist/src/cli.js plan ./my-site    # get the 3-path exit plan

Zero runtime dependencies. 100% local — your code never leaves your machine.

Quick example

$ picoexit audit ./my-site

# picoexit audit
- Pico delivery: **cdn** (v2.1.1)
- Pico selectors used: **45 distinct**, 85 occurrences

## Drift-risk selectors
- **role=search** (high risk): Pico styles [role=search] focus behavior
  with :has() selectors; core behavior, not just decoration.
  - index.html:29
...

Exit code: 1 (high drift-risk selectors in use)

audit exits non-zero when high drift-risk selectors are in use, so you can gate CI on it. Add --json for machine-readable output, --no-strict to always exit 0.

$ picoexit migrate ./my-site --to tailwind --dry-run

--- index.html
+++ index.html
@@ line 39 @@
- <div class="grid">
+ <div class="grid gap-4 md:grid-cols-3">

## Manual work (not guessed by the codemod)
- index.html:19 — `.dropdown`: CSS-only <details> dropdown; Bulma dropdown
  needs its JS trigger structure. Manual rebuild.

Re-run with --apply to write the changes. Anything without a verified mapping is listed as manual work — never guessed.

Commands

Command	What it does
picoexit audit <dir> [--json] [--no-strict]	Scan .html/.css/.js/.ts/.jsx/.tsx for Pico usage: delivery method (CDN / npm / vendored), version, every selector used with file:line, counts per component, drift-risk flags
picoexit plan <dir> [--json]	3-path exit plan (stay pinned / migrate / community fork) with checklists, codemod coverage per target, and manual-effort estimates
picoexit migrate <dir> --to tailwind|bulma|openprops [--dry-run] [--apply]	Codemod Pico classes to the target; dry-run prints a diff preview (default)

Mapping table

60 Pico v2.x hooks mapped to three targets. Target docs verified 2026-11-03 (Tailwind CSS v4, Bulma v1, Open Props). Highlights:

Pico	Tailwind v4	Bulma v1	Plain CSS + Open Props
.container	mx-auto w-full max-w-7xl px-4	container	max-width: var(--size-xxl); margin-inline: auto; ...
.grid	grid gap-4 md:grid-cols-N (N counted from your markup)	columns + column on children	display: grid; gap: var(--size-4);
article	rounded-lg border ... shadow-sm	box	surface / radius / shadow tokens
.contrast / .secondary / .outline	bg-zinc-900 ... / bg-zinc-200 ... / bg-transparent border-2	is-dark / is-light / is-outlined	gray tokens
.striped (table)	manual CSS	is-striped	tbody tr:nth-child(even) rule
dialog / modal classes	manual (replace with target's dialog controller)	manual	manual
data-theme / data-tooltip	manual (target's dark-mode / tooltip pattern)	manual	manual

Mappings are starting-point approximations — review against your design tokens. The full table lives in src/mappings.ts.

Why this exists

Pico's maintainer archived the project because AI can now generate lightweight semantic HTML directly. npm and jsDelivr stay up, so nothing breaks today — but every Pico site now carries the same unanswered question: how much of my site actually depends on Pico, and what rots first? Generic codemod platforms (codemod.com, jscodeshift) have no Pico-specific transforms, and framework upgrade tools only migrate within their own framework. picoexit is the only tool that measures your Pico footprint first, then prices the way out.

Out of scope (v0.1.0)

• picoexit never tells you which path to take — it measures, you decide.
• No migration of Pico theme/Sass customizations (standard classes/components only).
• No realtime tracking of community forks (known forks listed with verification date).
• Pico v1.x: audit works, codemod targets v2.x.

Roadmap

• Publish to the npm registry (pending owner token — install from source until then)
• GitHub Action wrapper (picoexit audit as a CI gate)
• More codemod targets (UnoCSS, plain cascade layers)
• Fork-tracking: re-check known community forks and report activity
• Vite/Astro plugin for in-build auditing

Disclaimer

picoexit is a community-built migration helper for Pico CSS users. It is not affiliated with, endorsed by, or connected to the Pico CSS project or its maintainer. "Pico CSS" is used here under nominative fair use to describe compatibility. Per the Pico license terms, the Pico CSS name and logo are not covered by the MIT license — this project uses its own name and branding.

License

MIT — see LICENSE.