iai-personal-memory-engine
bycodeabra· 1 MCP server
English | 中文
How to install iai-personal-memory-engine?
codeabra/iai-personal-memory-engine
$git clone https://github.com/codeabra/iai-personal-memory-engineInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install iai-personal-memory-engine by running `git clone https://github.com/codeabra/iai-personal-memory-engine`, then use it for the current task and follow its documentation at https://github.com/codeabra/iai-personal-memory-engine.
README.md
| 1 | **English** | [中文](./README_zh-CN.md) |
| 2 | |
| 3 | <p align="center"> |
| 4 | <img src="logo.png" alt="iai-pme" width="600"> |
| 5 | </p> |
| 6 | |
| 7 | <h3 align="center">The best open-source personal memory engine for AI coding assistants.</h3> |
| 8 | <p align="center">Every claim ships with the harness that proves it. Run the benchmarks yourself.</p> |
| 9 | <p align="center"><b>Persistent memory for any MCP client.</b> Local, encrypted, verbatim recall via vector search + knowledge graph. MIT.</p> |
| 10 | |
| 11 | <p align="center"> |
| 12 | <img src="https://img.shields.io/badge/release-v2.4.1-1f6feb?style=flat-square" alt="Release v2.4.1"> |
| 13 | <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-1f6feb?style=flat-square" alt="License: MIT"></a> |
| 14 | <img src="https://img.shields.io/badge/python-3.11%20|%203.12-3776ab?style=flat-square&logo=python&logoColor=white" alt="Python 3.11 | 3.12"> |
| 15 | <img src="https://img.shields.io/badge/platform-macOS%20|%20Linux-555?style=flat-square" alt="Platform: macOS and Linux"> |
| 16 | <img src="https://img.shields.io/badge/Windows-beta-dbab09?style=flat-square&logo=windows&logoColor=white" alt="Windows: beta"> |
| 17 | <img src="https://img.shields.io/badge/engine-Rust%20native-dea584?style=flat-square&logo=rust&logoColor=black" alt="Rust-native engine"> |
| 18 | </p> |
| 19 | <p align="center"> |
| 20 | <img src="https://img.shields.io/badge/LongMemEval%20R%405-0.962-2ea043?style=flat-square" alt="LongMemEval R@5 0.962"> |
| 21 | <img src="https://img.shields.io/badge/Rescue%4010-1.000-2ea043?style=flat-square" alt="Rescue@10 1.000"> |
| 22 | <img src="https://img.shields.io/badge/at%20rest-AES--256--GCM-2ea043?style=flat-square" alt="AES-256-GCM"> |
| 23 | <img src="https://img.shields.io/badge/local--only-no%20telemetry-2ea043?style=flat-square" alt="Local only, no telemetry"> |
| 24 | <img src="https://img.shields.io/badge/MCP-compatible-8957e5?style=flat-square" alt="MCP compatible"> |
| 25 | <a href="https://glama.ai/mcp/servers/CodeAbra/iai-mcp"><img src="https://glama.ai/mcp/servers/CodeAbra/iai-mcp/badges/score.svg" alt="Glama MCP score"></a> |
| 26 | </p> |
| 27 | |
| 28 | --- |
| 29 | |
| 30 | # iai-pme |
| 31 | |
| 32 | **Your AI assistant forgets you every session. iai-pme gives it a memory that doesn't.** |
| 33 | |
| 34 | *Independent Autistic Intelligence — a personal memory engine. Fully local, ambient. Works with Claude Code, Claude Desktop, Cursor, Codex CLI, Gemini CLI, Cline, Continue.dev, Zed, Cherry Studio, Goose, Aider, Hermes, OpenClaw, Le Chat, Kimi — anything that speaks MCP-over-stdio.* |
| 35 | |
| 36 | ## Table of contents |
| 37 | |
| 38 | - [What it is](#what-it-is) |
| 39 | - [Quick start](#quick-start) |
| 40 | - [Usage](#usage) |
| 41 | - [How it works](#how-it-works) |
| 42 | - [Built our own](#built-our-own) |
| 43 | - [Benchmarks](#benchmarks) |
| 44 | - [Configuration](#configuration) |
| 45 | - [Doctor](#doctor) |
| 46 | - [Notes for AI assistants](#notes-for-ai-assistants-helping-with-installation) |
| 47 | - [Status and limitations](#status-and-limitations) |
| 48 | - [Compatibility](#compatibility) |
| 49 | - [About the name](#about-the-name) |
| 50 | - [Authors](#authors) |
| 51 | - [License](#license) |
| 52 | - [Contributing](#contributing) |
| 53 | |
| 54 | --- |
| 55 | |
| 56 | ## What it is |
| 57 | |
| 58 | A local server that speaks the [MCP protocol](https://modelcontextprotocol.io) and gives Claude, and any other MCP-compatible assistant, a long-term memory. It captures every turn of every session verbatim, organizes those captures over time into a personal map of who you are, and serves a small slice of relevant memory back at the start of each new conversation. You never have to say *"remember this"* or *"what did we say last time?"*. |
| 59 | |
| 60 | <p align="center"><img src="docs/assets/slides/slide-02.jpg" width="850" alt="iai-pme"></p> |
| 61 | |
| 62 | I built this for myself. It worked. I've been running it daily for months, and now I'm sharing it. The benchmarks were mostly for my own curiosity. I wanted to know if it actually works or if I'd just gotten used to it. |
| 63 | |
| 64 | Under the hood it's not a wrapper around someone else's vector store and graph library — the parts that matter are my own code: the storage engine, the community-detection algorithm, the hyperdimensional memory substrate, and a native engine that makes it fast. More on that in [Built our own](#built-our-own). |
| 65 | |
| 66 | And |
Related
6 picksType
- cortexScientifically-grounded persistent memory for Claude — local-first, hybrid retrieval, 72 references.Jul 2026
- mcphubA hub server for mcp serversJul 2026
- wecom-bot-mcp-serverWeCom Bot MCP Server - A Python server for WeCom (WeChat Work) bot following the Model Context Protocol (MCP)Jul 2026
- hypertool-mcpDynamically expose tools from proxied servers based on an Agent PersonaOct 2025
- reddit-mcpReddit MCP ServerDec 2025
- agent-bridgeBridge between Claude Code and Codex — bidirectional agent communication via MCP Channel + JSON-RPCJul 2026
