Foundkeep

Found it? Keep it. Save readable pages, screenshots, highlights, images, and stray ideas with their original source in a private, searchable library.

Get Foundkeep · Add to Chrome · Privacy & data

Start collecting

1. Create a Foundkeep account and save your recovery code.
2. Install Foundkeep from the Chrome Web Store. The Store release is version 1.0.0 and Chrome delivers approved updates automatically.
3. Click Connect Foundkeep in the dashboard, pin the extension, and save your first find.

New captures save in the browser first and sync to the connected account when automatic sync is enabled. Saved pages can include their readable text, headings, visited and canonical URLs, author, publisher, dates, language, lead image, favicon, capture method, timestamps, extraction status, and a content fingerprint. Foundkeep does not store raw page HTML.

Open the dashboard to search, browse, export, revoke browsers, and control capture methods, page extraction, note sources, popup layout, context menus, sync, OCR, summaries, and tags. These account preferences flow to connected browsers without reinstalling the extension.

Earlier local captures remain local until you explicitly import them. Local and cloud copies are separate; deleting one does not erase the other. Accounts include up to 1,000 captures and 200 MiB of content.

Before replacing an unpacked installation, connect it and sync any local-only captures you want to keep. Chrome gives the Store build separate local storage; synced captures remain in the same Foundkeep account and dashboard.

Full extension guide

Repository

• apps/extension — Chromium MV3 capture popup, readable page extraction, durable account-bound sync queue, and local IndexedDB library.
• apps/site — React/Next customer website: static public pages, server-rendered auth/library, account-scoped queries, Motion interactions and a nonmodal saved-item reader.
• apps/web — shared first-party assets, extension downloads, and the previous static website retained for compatibility and rollback.
• apps/backend — Bun/Hono customer accounts, private capture APIs, SQLite storage, and bounded organization worker.
• packages/shared — schemas used by server-backed clients.
• Legacy companion, desktop, and browser-control integrations remain available for existing installations but are outside the customer onboarding flow.

Develop and verify

bun install --frozen-lockfile --ignore-scripts
bunx playwright-core install chromium
bun test
bun run test:extension
bun run test:web
bun run test:customer

The customer integration test also needs Tesseract English OCR and util-linux (prlimit); see customer operations. Browser tests use temporary profiles and synthetic captures. CHROMIUM_PATH can select an installed Chromium executable.

Deployment and release guide

React customer website deployment and tests