asciimoo / hister

项目摘要

Hister is a privacy-focused personal search engine that indexes your web browsing history and local files, enabling you to search through your own content locally. It offers multiple access methods including a web interface, terminal TUI, command line, and MCP integration for AI assistants. The project emphasizes user privacy with no telemetry and supports optional semantic search through external embedding endpoints.

为什么值得关注

Hister stands out as a privacy-respecting, self-hosted alternative to centralized search platforms that gives users full control over their personal search data. Its integration with MCP enables AI assistants to search your personal browsing history, creating a unique bridge between personal knowledge management and AI workflows. The active development pace (17 releases in 6 months) and strong community adoption (6K+ stars) demonstrate real demand for personal search infrastructure.

优势

• Privacy-first design with no telemetry or mandatory cloud services
• Multiple client interfaces: web UI, TUI, CLI, and MCP integration for AI assistants
• Browser extension for automatic indexing of visited pages
• Full-text content indexing rather than just titles and URLs
• Active development with frequent releases and multi-platform support
• Simple zero-config installation with multiple deployment options (binary, Docker, Homebrew, Nix)

局限性

• No automated tests despite having CI pipeline
• Semantic search requires external embedding endpoint configuration
• Relatively young project with limited track record
• Limited documentation examples and no dedicated examples directory

使用场景

• Searching your own browsing history to rediscover previously visited pages
• Personal knowledge management and local file search
• Integrating personal search capabilities into AI assistants via MCP
• Privacy-conscious teams wanting internal search infrastructure without cloud dependencies