Linkit: a local-first AI bookmark manager that summarizes and searches your links

You save a link, tell yourself you'll read it later, and then you never find it again.

That loop ran for years in my own bookmark collection. I'd end up with 2,000+ saved links and a realistic estimate of how many I'd actually opened: maybe five. The problem was never storage. Browser bookmarks store a URL and nothing else. There's no way to search by what you actually remember, no way to skim a link without opening it, and no way to notice that three of your saved links are the same article.

So I built Linkit, a desktop bookmark manager focused on one idea: the bookmark should remember why you saved it.

What it actually does

AI summaries and key takeaways. Every saved link gets fetched, summarized, and broken into a few takeaways. You scan the summary instead of opening twelve tabs. You can connect any OpenAI-compatible API, or run a local model through Ollama.

Semantic search that matches on meaning. Keyword search fails the moment you don't remember the wording. "how to make large file transfer faster" should find a link whose title says "transfer optimization." Search here embeds your library and matches on vector similarity.

A knowledge graph. Your library renders as an interactive network, connected by shared tags, collections, and semantic similarity. It's the fastest way to spot which topics you keep circling and which areas you've never touched at all.

Duplicate detection. Exact URL matches, plus AI-detected near-duplicates where the same content lives at different URLs. Review and merge with one click.

Link health checks. A background scan finds 404s and silently changed pages, so dead links don't quietly rot.

Local-first is the point

All bookmarks live in a local SQLite database on your machine. No account required, works fully offline, no telemetry, no third-party data sharing. If you want multi-device sync, there's an optional Supabase backend with row-level security where every record is bound to your own auth.uid() — including the developers can't read your data. That sync is free too, with no paywall gating it.

How it's built

Go for the backend, Wails to ship it as a native desktop app, React for the UI. The AI layer is provider-agnostic — point it at OpenAI, DeepSeek, a local Ollama instance, or anything that speaks the OpenAI API shape. Cross-platform binaries ship through GitHub Releases and Homebrew.

Installing

brew install blue-idea/tap/linkit

Windows, macOS DMG, and Linux AppImage/DEB are on the releases page.

It's free and open source

100% free. No paywall, no ads, no account, no feature locked behind a subscription. If you've ever lost a link you swore you'd come back to, try it: https://github.com/blue-idea/linkit

If it helps, a star genuinely helps other people find it.