pagefind

Pagefind Search

Enterprise-grade site search with zero server overhead.
Pagefind Search brings the blazing-fast, Rust-powered
Pagefind engine to Drupal. Instead of running
and maintaining a search server like Solr or Elasticsearch, Pagefind builds a
highly compressed static index that is served as plain files and
queried entirely in the visitor's browser via WebAssembly — fetching only the
few kilobytes of index a given query actually needs.

The result is instant, typo-tolerant, faceted search that costs your server
nothing per keystroke, scales to tens of thousands of documents, and keeps
working even on fully static or CDN-fronted deployments.

Why Pagefind Search?

• No search server to run. No Solr, no Elasticsearch, no
  background daemon, no per-query backend load. The index is a set of static
  files.
• Fast by design. Searches run in-browser over a compressed
  index, downloading only the chunks a query needs — typically tens of
  kilobytes — with no round-trip per keystroke.
• Built for Drupal. A native bridge to Pagefind: view
  modes, a field formatter, blocks, Views integration, Drush commands, and a
  configuration UI — not a bolt-on.

Key features

• Expert field formatter. A single, unified formatter to
  mark fields as filters, boost their search weight, expose them as metadata,
  or make them sortable — directly from Manage display.
• Faceted listing pages. Build filterable, sortable,
  paginated archives, libraries and directories with the bundled display UI —
  list, grid and table layouts, configurable columns, sidebars and pagination
  styles.
• Featured results. Promote chosen content to the top of
  results for specific keywords.
• Synonyms. Index-time synonym and equivalence expansion so
  visitors find content regardless of the exact words they type.
• Multisite search collections. Aggregate and search across
  multiple sites.
• Modern, themeable UI. Built on Pagefind's web components
  (Shadow DOM compatible), with a keyboard shortcut, modal and inline display
  modes.
• Developer tooling. Drush commands for index builds and
  diagnostics, plus a per-node "Pagefind preview" tab for inspecting exactly
  what gets indexed.

How it works

1. Enable the Pagefind Index view mode on your content types
   and tag fields with the Pagefind Expert formatter.
2. Build the index from the dashboard, or with
   

   drush
       pagefind:index

   for larger sites.

3. Drop in the search block or visit /search — queries run
   instantly in the browser.

Requirements

• Drupal 10.3+ or 11
• PHP 8.3 or newer
• A private file system ($settings['file_private_path']) for
  staging and binary storage — the Pagefind binary is downloaded at runtime
  into the private filesystem and is never web-accessible.

Scale & limits

Pagefind Search is engineered for sites with up to roughly
50,000 indexed documents. Within that envelope it delivers
excellent search with zero server query load. Above ~5,000 documents, build the
index via Drush rather than the browser batch UI. For very large corpora where
fine-grained relevance tuning is critical, a server-side backend may still be
the better fit.

Security & quality

The module is held to a high engineering bar: PHPCS-clean against the Drupal
and DrupalPractice standards, PHPStan level 8, a comprehensive kernel and
functional test suite, and a dedicated security pass covering the binary
download (checksum-verified, hardened against URL injection), archive
extraction, access control and output escaping.

Credits

Pagefind Search is a Drupal integration for the open-source
Pagefind engine by CloudCannon (MIT). The
Drupal module is licensed GPL-2.0-or-later.