asciimoo/hister

What it solves

Hister provides a private, self-hosted search engine for your personal data. It solves the problem of forgetting where information was found across various web pages and local files by indexing their full contents for easy retrieval.

How it works

It functions as a local server that indexes full-text content from visited web pages (via browser extensions for Chrome and Firefox) and local files. Users can search this index through a web interface, a terminal UI, or a command-line interface. It also supports optional semantic search by connecting to a configurable embeddings endpoint to find documents by meaning rather than just keywords.

Who it’s for

Individuals who want a private way to organize and search their browsing history and local documents without relying on cloud services or telemetry.

Highlights

• Full-text indexing: Searches actual page and file contents, not just titles.
• Privacy-first: No telemetry or mandatory cloud services; runs on user-controlled infrastructure.
• AI Integration: Supports the Model Context Protocol (MCP) to allow AI assistants to search the index.
• Semantic Search: Optional capability to find documents based on meaning via external embeddings.
• Flexible Access: Searchable via web UI, terminal, and CLI.