Skip to content

Extract from any page, without uploading it.

Server-side URL fetchers miss anything rendered by JavaScript — and they require sending the page through someone's server. This site takes the opposite approach: a bookmarklet that scans the page you already have open, entirely on your device.

Last reviewed: September 16, 2026 · English only for now

Install the bookmarklet

  1. Show your bookmarks bar (Ctrl/Cmd + Shift + B in most browsers).
  2. Drag this link onto the bar: Extract emails ↓
  3. Open any page, click the bookmark — matches appear in a floating panel with a Copy button.

Prefer manual install? Create a bookmark, paste the link target of the button above into its URL field, and save.

How it stays private

  • Reads document.documentElement.outerHTML of the current tab only.
  • Decodes mailto: targets plus visible addresses; deduplicates case-insensitively.
  • Makes zero network requests — verify in DevTools like the on-site tools.
  • Asks for no permissions, stores nothing, survives no longer than the panel.

Full browser extension (reference design)

A Manifest V3 extension is the planned next step for one-click scanning without the bookmarks bar. The design constraint is fixed: the content script may read the active tab, but extracted text must never leave the device — export happens through the same local copy/download path as this site.

{
  "manifest_version": 3,
  "name": "Email Extractor — local page scan",
  "version": "1.0.0",
  "description": "Extract emails from the current page. 100% local, no servers.",
  "permissions": ["activeTab", "scripting", "clipboardWrite"],
  "host_permissions": [],
  "action": { "default_title": "Extract emails on this page" }
}
  • Permissions limited to activeTab, scripting and clipboardWrite — no host access, no background scraping.
  • Same matching rules as the on-site engine, so results agree between page and tool.
  • The publishable package (manifest, content script, popup, icons, store checklist) already ships in this site's repository under extension/ — store submission itself needs a human account and review.

Frequently asked questions

Does the bookmarklet upload the page anywhere?

No. It runs entirely in the current tab, scans the rendered DOM and mailto links, and shows results in a floating panel. The page content never leaves your device.

Why a bookmarklet instead of URL fetching?

A bookmarklet sees exactly what you see — including addresses rendered by JavaScript after load, which server-side fetchers miss. It also needs no permissions, no store install and no account.

What are its limits?

It reads one rendered page at a time, cannot reach behind logins it cannot see, and cannot read addresses baked into images. For files and bulk lists, use the on-site tools instead.

Open the Email ExtractorRead the trust center