Hister

Give Your AI Assistant a Private Memory | Hister

AI assistants are very good at explaining things, but they usually have a weak<br>memory of your own context.<br>They do not know which documentation page you read last week. They do not know<br>which article convinced you to try a library. They do not know which migration<br>guide you followed, which GitHub issue had the workaround, or which security<br>writeup you already trusted.<br>That context often lives in your browser history, bookmarks, local notes, and<br>half-remembered search queries.<br>Hister turns that context into a private search index. With MCP support, it can<br>also make that index available to AI assistants.<br>A Quick Introduction to Hister<br>Hister is a personal search engine for pages and files you care about.<br>The browser extension automatically indexes pages as you browse. The command line<br>tool can import browser history, crawl documentation sites, index URLs manually,<br>and add local files. The web interface lets you search everything with full-text<br>search, field filters, labels, facets, priority results, and optional semantic<br>search.<br>The important part is where the data lives.<br>Your browsing history, indexed pages, search queries, and local files stay in<br>your Hister instance. For most people, that means they stay on the same machine<br>or on a server they control.<br>This also means you do not need to set up a separate assistant skill for every<br>site you use. You do not need to give your assistant individual API access to<br>GitHub, documentation sites, issue trackers, forums, wikis, and other sites<br>that require authentication. If the useful page is already in Hister, the<br>assistant can search Hister instead.<br>Instead of wiring every private service into your assistant, Hister gives it<br>one search interface over material you already indexed.

That makes Hister useful as a daily search tool by itself. MCP adds another<br>layer: an assistant can ask Hister for relevant pages instead of guessing from<br>general model knowledge or fetching random web results.<br>Practical Hister MCP Workflows<br>Here are a few ways Hister MCP can be useful in everyday work. The common<br>pattern is simple: the assistant asks Hister first, then works from pages and<br>files that are already part of your own index.<br>Find the Article You Remember Vaguely<br>You know the situation: you read a good explanation of a bug, a design pattern,<br>or a deployment problem, but you cannot remember the title or the site.<br>Without Hister, you ask an assistant a broad question and get a broad answer.<br>It might be helpful, but it is not grounded in the source you remember.<br>With Hister MCP configured, you can ask:<br>Search my Hister index for the article I read about PostgreSQL migration<br>locking and summarize the most relevant result. The assistant can search Hister for PostgreSQL migration locking, inspect the<br>stored result, and answer based on the pages in your own index.<br>If semantic search is enabled on your Hister server, the assistant can also ask<br>for semantic matching. That helps when you remember the idea but not the exact<br>words. If semantic search is not available on the server, Hister falls back to<br>normal keyword search.<br>Explain Code with Documentation You Already Indexed<br>AI coding assistants often answer from broad training data. That is useful, but<br>it can be wrong for the exact version of a library or framework you use.<br>Hister gives the assistant a more specific source of context.<br>You can crawl the documentation for a project:<br>hister index --recursive<br>--allowed-domain=docs.example.com<br>--max-depth=4<br>https://docs.example.com/ You can also start from Hister’s prepared documentation datasets instead of<br>crawling everything yourself. The datasets page includes imports<br>for common reference material such as language and platform documentation, so<br>your assistant can search those docs through Hister immediately after import.<br>Then, inside your AI assistant, ask:<br>Using my Hister index, find the official docs for configuring connection<br>timeouts in this library. Then explain which option applies to this code. The assistant can search your indexed docs, retrieve previews, and use those<br>pages when explaining the code.<br>This is especially useful for internal documentation, private wikis, older API<br>versions, or documentation that is hard to find through a normal web search.<br>Turn Research into a Source-Backed Brief<br>When you research a topic over several days, the useful material usually ends<br>up scattered across tabs and history.<br>With Hister, browsing becomes capture. With MCP, the assistant can turn that<br>captured material into a brief.<br>Example prompt:<br>Search my Hister history for pages about supply chain attacks from the last<br>month. Group the findings by incident, list the affected projects, and cite the<br>pages you used. The assistant can use date filters in the Hister search tool, then retrieve<br>stored previews for the most relevant pages. The result is not just a generic<br>summary of supply chain attacks. It is a summary of the material you...