Hister is a private personal search engine for web pages and local files, designed to let you find content again from the browser, terminal, or an AI assistant via MCP.
Hister is a self-hosted personal search engine built in Go. It indexes the full contents of pages you visit and files you keep, so you can search across material you have already encountered. The project also provides access through a web interface, terminal tools, and MCP-based AI integrations.
It addresses the common problem of losing track of information you have already seen on the web or stored locally. Instead of relying on title-based browser history or a general-purpose search provider, Hister is meant to help you search the actual content of pages and files you chose to keep.
Conceptually, Hister collects content from sources you enable, builds a local or self-hosted index, and then lets you query that index with search features such as phrases, filters, wildcards, negation, aliases, and result prioritization. The README also says it can optionally support semantic search through a configured embeddings endpoint, and it can ingest content from browser history, bookmarks, local directories, and imported files.
It is gaining attention because it sits at the intersection of personal search, privacy, and AI-assisted workflows. The repository explicitly highlights local-first operation, browser extensions for Firefox and Chrome, MCP support for AI assistants, and multiple installation paths such as direct downloads, Homebrew, Docker, and Nix, which lowers the barrier to trying it.
Based on the README, the closest comparisons are traditional web search engines, browser history, browser bookmarks, and general note or file search tools, since Hister is positioned as a private alternative to rediscovering content you already have. The README does not name direct competitors, so it is safest to say it overlaps with personal search engines, local document search, and browser-based history tools.
AI-explained · grounded in each repo's README