asciimoo/hister: Your personal search engine · GitHub

Your personal search engine

Hister is a personal search engine for the pages you go to and the information you retain. It indexes their full contents so you will discover data once more from the online interface, terminal, or an AI assistant related via MCP.

Try the demo · Download Hister · Read the quickstart · Documentation

1. Download the binary in your platform from the latest release, then rename it to hister (hister.exe on Windows).

2. On Linux or macOS, make it executable:

3. Start Hister on Linux or macOS:

   On Windows, run .hister.exe hear in PowerShell.

   Keep this terminal open whereas utilizing Hister. The server should be operating to index pages and search them.

4. Open http://127.0.0.1:4433 and set up the browser extension for Firefox or Chrome.

5. Visit an online web page with the extension enabled, then return to Hister and seek for a phrase from that web page to search out your first listed consequence.

No configuration is required for a neighborhood private setup. See the complete quickstart to decide on what Hister indexes.

To search current content material, import browser history, index local directories, or import files.

Alternative set up strategies embody Homebrew (brew set up hister), Docker, and Nix. See the installation guide for directions.

• Privacy targeted: No telemetry or necessary cloud service. Run Hister regionally or on utilities you management.
• Full textual content indexing: Search the precise contents of visited pages and native information, not solely titles and URLs.
• Automatic browser indexing: Save newly visited pages with the Firefox or Chrome extension.
• Powerful queries: Use area filters, phrases, wildcards, negation, aliases, and consequence priorities.
• Optional semantic search: Find paperwork by that means via an embeddings endpoint you configure.
• Crawler and browser import: Index web sites or usher in current browser historical past.
• Web, terminal, and MCP purchasers: Search from the browser, TUI, command line, or an AI assistant.
• Multi consumer assist: Keep every consumer’s paperwork and search outcomes separate on a shared server.

By default, Hister has no telemetry and no cloud sync. The browser extension sends listed web page content material solely to the Hister server you configure, other than downloading web page favicons. The server shops paperwork and search indexes on that server.

Optional semantic search sends doc textual content to the embeddings endpoint you select. Review the privacy overview and semantic search configuration earlier than enabling distant integrations.

Unlike conventional serps, Hister builds a private search index from the online pages and information you select to maintain. Your content material stays in your Hister server, making it helpful for locating data you’ve got already encountered with out counting on a third-party search supplier.

Requirements are Go 1.26, npm, and a C compiler for CGO dependencies.

git clone https://github.com/asciimoo/hister.git
cd hister
./handle.sh construct

To work on the internet app with sizzling reload and automated Go rebuilds:

This begins a Vite growth server and the Go backend with automated rebuilds via air.

Join us on IRCNet in #hister or on Discord.

Read CONTRIBUTING.md earlier than submitting a change. Bugs and options belong within the issue tracker. For safety experiences, see SECURITY.md.

AGPLv3 or any later model