Use Hutch from
Claude Code.
Persistent memory for your terminal coding sessions.
Claude Code lives in your terminal, but its memory dies with each session. Hutch fixes that: ask Claude to drop bugs, todos, decisions, or docs-debt into Hutch, and they survive across days, repos, and machines.
Hutch installs as an HTTP MCP server. Install is one command, and OAuth handles auth, so there are no API keys to manage.
Install
Fastest: paste this into Claude Code and it sets itself up. You approve the sign-in in your browser.
Connect Hutch (hutchdb.com) to Claude Code as an MCP server.
- URL: https://app.hutchdb.com/api/mcp
- Transport: remote streamable HTTP
- Auth: OAuth (no API key). I'll approve in the browser.
How to do it in Claude Code:
- Run: claude mcp add --transport http --scope user hutch https://app.hutchdb.com/api/mcp
- Then run claude mcp login hutch, or tell me to run /mcp and authenticate hutch.
1. Add the server for my user account, not just this project, so it works everywhere.
2. Start the OAuth login, or tell me exactly what to click if you can't start it yourself.
3. Tell me if I need to restart or reload you. Then call hutch_list_collections to confirm it works.Or set it up by hand:
01Step 1. Connect
Run this in your terminal:
claude mcp add --transport http --scope user hutch https://app.hutchdb.com/api/mcp--scope user makes Hutch available in every project. Drop it to add Hutch to the current project only.
02Step 2. Authenticate
Start Claude Code and run:
/mcpSelect hutch and authenticate. A browser window opens: sign in with GitHub or Google and approve. For headless use, run claude mcp login hutch in your shell first.
03Step 3. Try it
From a Claude Code session, try:
- >"Store this article in my Hutch reading-list collection."
- >"Save the bugs we just discussed to a "tickets" collection in Hutch."
- >"What collections do I have in Hutch?"
- >"Find my notes about caching in Hutch."