ReMe

ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.

What is ReMe?

ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.

Overview of ReMe

What it is

ReMe is a local-first, self-evolving personal knowledge base for AI agents. It stores durable memory as ordinary Markdown files with frontmatter and wikilinks. Users and agents can inspect, edit, move, sync, and back up these files with familiar tools. Indexes and generated metadata remain rebuildable. ReMe progressively turns conversations and resources into daily notes and long-term knowledge, preserving sources while refining facts, preferences, procedures, and relationships over time. It uses BM25, optional embeddings, and wikilink expansion for precise and context-aware recall. One memory workspace can be shared across agents such as personal assistants, coding agents, and other agent runtimes through native integrations, SKILL.md, CLI, HTTP, MCP, or Python APIs.

Key features

  • Memory as File, File as Memory: durable memory stored as ordinary Markdown with frontmatter and wikilinks.
  • Self-evolving knowledge base: progressively turns conversations and resources into daily notes and long-term knowledge.
  • Recall is precise and context-aware: uses BM25, optional embeddings, and wikilink expansion for line-level passage retrieval.
  • One memory workspace works across agents: personal assistants, coding agents, and other runtimes can share the same local workspace.
  • Accessible via CLI, HTTP API, MCP server, or embedded Python API.
  • Optional decision-model context compaction and multimodal memory support (with specific models).

Use cases

  • Building a personal knowledge base for AI agents that persists across sessions.
  • Enabling multiple agents (e.g., personal assistants, coding agents) to share and evolve a common memory workspace.
  • Converting conversations and resources into structured, searchable Markdown notes.
  • Providing agents with precise recall of past interactions and learned facts without loading the entire knowledge base.

Getting started

ReMe requires Python 3.11+. Install from pip: bash pip install 'reme-ai[core]' Install from source: bash git clone https://github.com/agentscope-ai/ReMe.git cd ReMe pip install -e reme_studio -e '.[core]' cd reme_studio npm ci npm run build:static cd .. Docker (requires Docker and Compose 2.24.0+): bash mkdir -p .reme docker compose up --build -d Start the service: bash reme start Default service address is 127.0.0.1:2333. See the repository README for environment variables and configuration.

License and pricing

ReMe is open source under the Apache-2.0 license. No pricing information is provided.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/agentscope-aireme.svg)](https://yunodb.com/tools/agentscope-aireme?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/agentscope-aireme?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/agentscope-aireme.svg" alt="Featured on YunoDB" height="20"></a>