Find content from visited web pages and local files on your own machine
When you cannot remember a specific sentence from a document you read yesterday or a news article from last week, you do not need to dig through browser history or rely on external search engines. Nogon indexes the full content of web pages you have visited and local files, allowing you to find that content immediately through a web interface, terminal, or AI assistant. It is a personal search tool that does not depend on third-party search services and keeps your data on your server.
From Hister to Nogon: Installation and Basic Operation
This project was previously known as Hister and is now named Nogon. You can start the server by downloading the binary for your platform, granting execute permissions, and running the nogon listen command. Page indexing and search are only available while the server is running, so you must keep the terminal open. No default configuration is required, and it can be used immediately in a local personal environment without additional setup. Installing the Firefox or Chrome browser extension automatically indexes newly visited pages.
Searching by Body Content, Not Just Titles
Nogon indexes the actual text content of visited pages and local files, not just titles or URLs. This allows you to use advanced query features such as field filters, syntax, wildcards, negation conditions, aliases, and result prioritization. You can perform searches via a web browser, terminal user interface (TUI), command line, or an AI assistant connected through MCP. It also supports importing existing browser history or bookmarks and indexing local directories. In multi-user environments, documents and search results for each user are managed separately.
Where Data Goes: Privacy and Semantic Search
By default, Nogon has no telemetry or cloud synchronization features. The browser extension sends page content to be indexed only to the Nogon server configured by the user and does not send data externally, except for downloading page favicons. If you optionally enable semantic search, document text is sent to the embedding endpoint specified by the user. It is recommended to review the privacy overview and semantic search settings before enabling remote integration. All data is stored on the Nogon server and does not rely on third-party search providers.
Development Environment and License
Development requires Go 1.26, npm, and a C compiler for CGO dependencies. You can build the project using the ./manage.sh build command, and when developing the web app, you can run the Vite development server and automatic rebuilding of the Go backend via npm run serve:app. This project follows the AGPLv3 or later license. If you have plans for distribution or commercial use, you must check the terms of that license. Additionally, if you use semantic search, you should review your security policies considering that data is sent to external embedding services.