hister: A personal search engine that digs into visit history and local files

This is a personal search engine that indexes the full body text of visited web pages and local files. Unlike traditional methods that match only titles or URLs, it retrieves information based on the actual text within pages. It operates completely independently on local or self-hosted infrastructure without telemetry or cloud synchronization.

Newly visited pages are automatically saved via Firefox or Chrome extensions. It supports advanced query syntax, including field filters, phrases, wildcards, and negation conditions, and optionally enables semantic search through configured embedding endpoints. It also offers features to import existing browser history or directly index websites using a crawler.

The same index can be searched via the web interface, terminal, command line, and AI assistants connected through MCP. In multi-user environments, each user's documents and search results are managed separately. Combining a backend written in Go with a web UI, it is suitable for developers who prioritize personal data sovereignty and users building AI workflows.