FluctlightDB

Guides

How to point an assistant at a brain

Guides are the steps from the rest of this site, in order. The first one is the MCP server for Claude and Cursor. It does not add a claim the integration page does not already make.

On this page

  • Persistent memory for Claude and Cursor

    The repository ships an MCP server so Cursor, Claude and Codex can call memory_remember and memory_recall on a brain directory on your machine. These steps are the ones on the MCP integration page, in order. Claude and Cursor were not attached, and a server process was not left listening.

What a guide will not pretend

If a snippet was not run, the guide says so. If a release note says a fix shipped, the guide says the note says that, and it does not name a version to install. Offline recall is still lexical unless you bring an embedder.