Skip to content
Back to search

membase

Memory layer for AI agents — SDKs, CLI and MCP server for Membase

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/unibaseio/membase with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/unibaseio/membase'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/unibaseio--membase/badge.svg)](/listings/unibaseio--membase)

HTML

<a href="/listings/unibaseio--membase"><img src="/api/listings/unibaseio--membase/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

<p align="center"> </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](https://docs.membase.ai/build/reference/sdk-quickstart/). The Python package is `unibaseio-membase` (import `membase`); `membase` on PyPI is unrelated.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 153 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/membase-logo.png" width="72" alt="Membase">
  3. </p>
  4. <h1 align="center">Membase</h1>
  5. <p align="center">
  6. <b>One memory. Every AI. Only yours.</b><br>
  7. Teach it once. Every AI remembers.
  8. </p>
  9. <p align="center">
  10. <a href="https://pypi.org/project/unibaseio-membase/"><img src="https://img.shields.io/pypi/v/unibaseio-membase?label=pypi" alt="PyPI"></a>
  11. <a href="https://www.npmjs.com/package/@unibaseio/membase"><img src="https://img.shields.io/npm/v/@unibaseio/membase?label=npm" alt="npm"></a>
  12. <a href="https://github.com/unibaseio/membase/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue" alt="MIT"></a>
  13. </p>
  14. <p align="center">
  15. <a href="https://membase.com">Website</a> ·
  16. <a href="https://docs.membase.com">Docs</a> ·
  17. <a href="https://app.membase.com">Web app</a> ·
  18. <a href="https://discord.gg/nB9EfPGsSf">Discord</a> ·
  19. <a href="https://x.com/Unibase_AI">X</a>
  20. </p>
  21. <p align="center">
  22. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/hero.svg" width="840" alt="Add sources, manage one memory, use it from every AI and app">
  23. </p>
  24. Membase is the memory layer for AI agents. It keeps one memory for everything you and your agents
  25. know, and every AI you use reads it.
  26. You stop re-explaining yourself each time you switch models or tools.
  27. - **Add.** Facts, documents, web pages and chat exports from Claude or ChatGPT go into one memory.
  28. - **Manage.** Memories keep projects apart; search them, ask them questions, forget what should go.
  29. - **Use.** Your code, Claude Code, Cursor, Codex and ChatGPT read the same memory through the SDK or MCP.
  30. This repository is the developer side: Python and TypeScript clients, the `membase` command line
  31. and an MCP server. The memory lives in your Membase account, which the
  32. [web app](https://app.membase.com) and browser extension also fill, or on your machine with
  33. `unibaseio-membase[local]`.
  34. ## Write once, read from any AI
  35. <p align="center">
  36. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/in-action.svg" width="840" alt="Your app writes a memory; any AI reads it, here Claude Code">
  37. </p>
  38. Recordings: [Python SDK](https://github.com/unibaseio/membase/blob/main/assets/videos/sdk.mp4) (57 s) ·
  39. [Claude Code](https://github.com/unibaseio/membase/blob/main/assets/videos/switch-ai.mp4) (1 min 33 s). The answer above is illustrative.
  40. ## Accuracy
  41. <p align="center">
  42. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/results.svg" width="840" alt="LoCoMo 93.1%, LongMemEval_S 92.6%, DMR 92.2%">
  43. </p>
  44. 93.1% on LoCoMo, 92.6% on LongMemEval_S and 92.2% on DMR, using about 3× fewer tokens than the full
  45. history on LoCoMo and 11× fewer on LongMemEval_S. Method and results: [membase.com/research](https://membase.com/research)
  46. ## Quickstart
  47. <p align="center">
  48. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/architecture.svg" width="840" alt="Hosted, local in process, or local over HTTP">
  49. </p>
  50. ```bash
  51. pip install unibaseio-membase # hosted · Python 3.10+ · needs MEMBASE_API_KEY
  52. pip install 'unibaseio-membase[local]' # local · Python 3.12 or 3.13 · needs OPENAI_API_KEY
  53. pip install 'unibaseio-membase[media]' # local + photos, videos and audio
  54. npm install @unibaseio/membase # TypeScript · Node 18+
  55. npm install -g @unibaseio/membase-cli # the `membase` command (hosted) without Python
  56. ```
  57. ```python
  58. from membase import Membase
  59. m = Membase() # hosted
  60. # m = Membase(local=True) # local, in ~/.membase
  61. m.memories.add("We picked Postgres for the ledger service.") # a fact
  62. m.search("what database is the ledger on?") # {"items": [...], "partial": [...]}
  63. m.add("Design notes …", custom_id="design-1") # a document
  64. m.profile_facts.list() # who the user is (Your profile ticked)
  65. m.capabilities() # what this key may do, and the rules
  66. ```
  67. No call names a Memory: reads cover every Memory the key reaches, and writes land in its write
  68. target (a key that reaches several Memories and has none is refused with `WriteTargetRequiredError`).
  69. Ids are typed (`mem_…` a fact, `doc_…` a document, `med_…` a photo or video), lists are
  70. `{items, has_more, partial}`, and a removal without `confirm=True` raises `ConfirmationRequiredError`.
  71. Local memory answers the same routes with the same shapes.
  72. Every method, its REST route and MCP tool, and the access it needs: [SDK reference](https://docs.membase.com/build/reference/sdk-quickstart/).
  73. Upgrading from 0.3: see the [changelog](https://github.com/unibaseio/membase/blob/main/CHANGELOG.md).
  74. The Python package is `unibaseio-membase` (import `membase`); `membase` on PyPI is unrelated.
  75. ## Command line
  76. ```bash
  77. membase --local add "We picked Postgres for the ledger"
  78. membase --local search "what did we pick for the ledger?"
  79. membase --local memories list
  80. membase --local ask "Which database is the ledger on?"
  81. membase --local import ~/Downloads/claude-export.json # Claude, ChatGPT, markdown or JSON chats
  82. membase --local agent ingest trace.json --agent coder
  83. membase --local agent search "fix flaky deploy" --agent coder
  84. membase --local media add ~/Movies/trip.mp4 # shots, speech and speakers
  85. membase --local media search "someone skiing" # any language
  86. ```
  87. Without `--local` (or `--store DIR`, both before the command) commands use the hosted API; `agent`,
  88. `import` and `serve` need `unibaseio-membase[local]`, and `media add` and `media ask` (local only)
  89. need `unibaseio-membase[media]`: see [Media memory](docs/reference.md#media-memory). A local store
  90. with several Memories writes where `MEMBASE_WRITE_TARGET` names.
  91. ## MCP
  92. <p align="center">
  93. <img src="https://raw.githubusercontent.com/unibaseio/membase/main/assets/mcp.svg" width="840" alt="MCP clients, hosted or local server, same tools">
  94. </p>
  95. ```bash
  96. claude mcp add membase -- membase --local mcp # local
  97. claude mcp add --transport http membase https://api.app.membase.com/mcp-http \
  98. --header "Authorization: Bearer $MEMBASE_API_KEY" # hosted
  99. ```
  100. Hosted without the header, the connection signs you in and is read-only. Setup for Claude, ChatGPT,
  101. Cursor, Codex and other clients: [Connect your AI](https://docs.membase.com/connect/).
  102. Gemini CLI, as an extension (it signs you in on first use; `/mcp auth membase` signs in again):
  103. ```bash
  104. gemini extensions install https://github.com/unibaseio/membase
  105. ```
  106. Cline: add the server to `cline_mcp_settings.json`; [llms-install.md](llms-install.md) has the
  107. entry, with and without a developer key.
  108. ## Docs
  109. - [SDK reference](https://docs.membase.com/build/reference/sdk-quickstart/): methods, REST routes, MCP tools, access levels
  110. - [Memory operations](https://docs.membase.com/build/guides/memory-operations/): Memories, facts, documents, media, profile, ask
  111. - [Authentication](https://docs.membase.com/build/reference/authentication/): keys and what each one can reach
  112. - [Connect your AI](https://docs.membase.com/connect/): per-client MCP setup
  113. - [Agent framework integrations](integrations/): LangChain, Microsoft Agent Framework, Dify, Hermes Agent, OpenClaw and Strands Agents
  114. - [Local engine](https://github.com/unibaseio/membase/blob/main/docs/reference.md): what it stores, `membase serve`, settings
  115. ## License
  116. MIT; see [LICENSE](https://github.com/unibaseio/membase/blob/main/LICENSE). `unibaseio-membase[local]` also
  117. installs unibaseio-membase-core, the compiled engine, under its own Unibase license.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.