membase

<p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/membase-logo.png" width="72" alt="Membase"> </p> <h1 align="center">Membase</h1> <p align="center"> <b>One memory. Every AI. Only yours.</b><br> Teach it once. Every AI remembers. </p> <p align="center"> <a href="https://pypi.org/project/unibaseio-membase/"><img src="https://img.shields.io/pypi/v/unibaseio-membase?label=pypi" alt="PyPI"></a> <a href="https://www.npmjs.com/package/@unibaseio/membase"><img src="https://img.shields.io/npm/v/@unibaseio/membase?label=npm" alt="npm"></a> <a href="https://github.com/unibaseio/membase-ai/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-proprietary-blue" alt="Proprietary"></a> </p> <p align="center"> <a href="https://membase.ai">Website</a> · <a href="https://docs.membase.ai">Docs</a> · <a href="https://www.app.membase.ai">Web app</a> · <a href="https://discord.gg/nB9EfPGsSf">Discord</a> · <a href="https://x.com/Unibase_AI">X</a> </p> <p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/hero.svg" width="840" alt="Add sources, manage one memory, use it from every AI and app"> </p>

Membase is the memory layer for AI agents. It keeps one memory for everything you and your agents know, and every AI you use reads it. You stop re-explaining yourself each time you switch models or tools.

• Add. Facts, documents, web pages and chat exports from Claude or ChatGPT go into one memory.
• Manage. Containers keep projects apart; search it, ask it questions, forget what should go.
• Use. Your code, Claude Code, Cursor, Codex and ChatGPT read the same memory through the SDK or MCP.

This repository is the developer side: Python and TypeScript clients, the membase command line and an MCP server. The memory lives in your Membase account, which the web app and browser extension also fill, or on your machine with unibaseio-membase[local].

Write once, read from any AI

<p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/in-action.svg" width="840" alt="Your app writes a memory; any AI reads it, here Claude Code"> </p>

Recordings: Python SDK (57 s) · Claude Code (1 min 33 s). The answer above is illustrative.

Accuracy

<p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/results.svg" width="840" alt="LoCoMo 93.1%, LongMemEval_S 92.6%, DMR 92.2%"> </p>

93.1% on LoCoMo, 92.6% on LongMemEval_S and 92.2% on DMR, using about 3× fewer tokens than the full history on LoCoMo and 11× fewer on LongMemEval_S. Method and results: membase.ai/research

Quickstart

<p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/architecture.svg" width="840" alt="Hosted, local in process, or local over HTTP"> </p>

pip install unibaseio-membase              # hosted · Python 3.10+ · needs MEMBASE_API_KEY
pip install 'unibaseio-membase[local]'     # local  · Python 3.12 or 3.13 · needs OPENAI_API_KEY
npm install @unibaseio/membase              # TypeScript · Node 18+
npm install -g @unibaseio/membase-cli       # the `membase` command (hosted) without Python

from membase import Membase

m = Membase()                    # hosted
# m = Membase(local=True)        # local, in ~/.membase

m.memories.add("We picked Postgres for the ledger service.", container="Engineering")
m.search("what database is the ledger on?")
m.add("Design notes …", container="Engineering", custom_id="design-1")   # a document
m.ask("Which database did we choose for the ledger?")

Every method, its REST route and MCP tool, and the access it needs: SDK reference. The Python package is unibaseio-membase (import membase); membase on PyPI is unrelated.

Command line

membase --local add "We picked Postgres for the ledger" --container Engineering
membase --local search "what did we pick for the ledger?"
membase --local ask "Which database is the ledger on?"
membase --local import ~/Downloads/claude-export.json     # Claude, ChatGPT, markdown or JSON chats
membase --local agent ingest trace.json --agent coder
membase --local agent search "fix flaky deploy" --agent coder

Without --local (or --store DIR, both before the command) commands use the hosted API; agent, import and serve need unibaseio-membase[local].

MCP

<p align="center"> <img src="https://raw.githubusercontent.com/unibaseio/membase-ai/main/assets/mcp.svg" width="840" alt="MCP clients, hosted or local server, same tools"> </p>

claude mcp add membase -- membase --local mcp                                   # local
claude mcp add --transport http membase https://api.app.membase.ai/mcp-http \
  --header "Authorization: Bearer $MEMBASE_API_KEY"                             # hosted

Hosted without the header, the connection signs you in and is read-only. Setup for Claude, ChatGPT, Cursor, Codex and other clients: Connect your AI.

Docs

• SDK reference: methods, REST routes, MCP tools, access levels
• Memory operations: containers, documents, memories, profile, ask
• Authentication: keys and what each one can reach
• Connect your AI: per-client MCP setup
• Local engine: what it stores, membase serve, settings

License

Copyright © 2026 Unibase. All rights reserved. Proprietary; see LICENSE. unibaseio-membase[local] also installs unibaseio-membase-core, the compiled engine, under its own Unibase license.