Hister: Self-Hosted Private Search Engine
/hister
Hister is a self-hosted search engine that indexes the pages you browse and the files you keep, searchable from a web UI, a terminal client, or an MCP-connected AI assistant. Reach for it if you want full-text search across your own browsing history and documents without sending anything to a third party. Skip it if you need search across a team's shared drive rather than one person's history and files.
What is Hister?
Hister is a private search engine, written in Go, that builds a full-text index of pages you visit and local files you point it at. A Firefox or Chrome extension captures pages as you browse, while a crawler and history import bring in what you already have. Content is searchable from the web interface, a terminal client, or through MCP, with an optional semantic search layer for meaning-based results.
Key Features of Hister
- ✓Full-text indexing of both visited pages and local files, not just their titles or URLs.
- ✓Automatic browser indexing via the Firefox and Chrome extensions, so newly visited pages get saved as you browse.
- ✓Query syntax with field filters, phrases, wildcards, negation, aliases, and result priorities for narrowing search results.
- ✓Optional semantic search that finds documents by meaning, provided you point Hister at an embeddings endpoint.
- ✓A crawler plus browser-history import for pulling in existing history or indexing whole sites.
- ✓Three client surfaces — a web interface, a terminal (TUI/CLI) client, and an MCP server for AI assistants.
- ✓Multi-user support is provided, ensuring that on a shared server, each user's documents and search results are kept separate.
Getting Started with Hister
Download the binary for your platform from the latest GitHub release, then rename it to hister (hister.exe on Windows). On Linux or macOS, run chmod +x hister to make it executable, then start it with ./hister listen; on Windows, run .\hister.exe listen in PowerShell. Open http://127.0.0.1:4433 in your browser and install the Firefox or Chrome extension. The README says no configuration is required for a local personal setup — the full quickstart at hister.org covers importing existing browser history and choosing what gets indexed.
Strengths
- ✓No telemetry or mandatory cloud service by default — the README says indexed content only goes to the Hister server you configure.
- ✓Query language goes beyond plain keyword matching, with field filters, phrases, wildcards, negation, aliases, and result priorities.
- ✓Three ways in — web UI, terminal client, and MCP — cover browsing, scripting, and AI-assistant workflows without separate tools.
- ✓Ships as a single Go binary with no required configuration for a local, single-user setup.
Considerations Before Using Hister
- △Multi-user and semantic search both mean running and securing your own server long-term — semantic search also sends document text to whatever embeddings endpoint you configure, so it's not private by default the way the base install is.
- △To compile from source, the Go language version 1.26 is required, as is the npm package manager and a C compiler to manage CGO dependencies; this isn't a simple, single-command process if you don't possess a Go toolchain already.
- △The project has a small but active user base, so don't expect the plugin ecosystem or community coverage of a mainstream search tool.
- △The README doesn't document upgrade paths, index size limits, or how it performs once history and file indexes get large.
Other Private Search Options
Common Questions About Hister
Hister is licensed under AGPL-3.0, per its LICENSE file, a copyleft license that requires anyone running a modified version as a network service to share their source changes too.
Hister has no telemetry and no cloud sync by default. The browser extension only sends indexed page content to the Hister server you configure, aside from downloading page favicons.
Hister ships official extensions for Firefox and Chrome, listed on the Mozilla Add-ons site and the Chrome Web Store, which capture pages automatically as you browse.
Hister can index local files on your computer, it indexes their full contents alongside visited pages, so both show up in the same search.
Hister supports multiple users on one shared server, keeping each user's documents and search results separate rather than pooling everyone's index together.
To build Hister from source, Go 1.26 and npm are necessary, along with a C compiler to handle CGO dependencies; afterward, clone the repo and run ./manage.sh build.
The problem it solves
Browser history search in Firefox or Chrome only matches page titles and URLs, so a page you skimmed last month and now can't recall the name of is effectively lost — and none of the built-in tools touch the PDFs, notes, or other local files sitting next to that browsing. Hister closes that gap by indexing the actual text of both visited pages and local files in one place you control, instead of your history living in the browser and everything else scattered across folders and other apps.
Best use cases
- •Finding a page you read weeks ago by a phrase you remember from its body text, not its title.
- •Building a personal knowledge base by combining crawled sites, imported browser history, and local files behind one search box.
- •Giving an AI assistant a way to search your own browsing and documents through the MCP server, instead of asking it to guess.
- •Running a small shared search server for a household or team via multi-user support, with each person's documents kept separate.
Who should try it — and who should skip
Try Hister if you're comfortable running a small Go server, want full-text search over your own browsing and local files, and would rather keep that index on hardware you control than in a browser's built-in history search. Skip it if you just want quick keyword lookup in your browser's existing history panel, or if you need a search tool for a team's shared file server rather than one person's history and documents — Hister's multi-user mode keeps things separate per user, it isn't built as shared team search.
Related repositories
Is hister worth your time?
ChatGPT, Claude and Perplexity can all read this page. Ask one of them what it makes of hister.