Repolode
felinics

felinics/Memoh

Open-source multi-agent platform that gives each AI agent its own cloud workspace with filesystem, desktop, browser, network and long-term memory.

Stars
2,721
Forks
258
Open issues
64
Last commit
5 hours ago

What is it?

Memoh runs AI agents in isolated cloud workspaces, so each agent has its own filesystem, desktop, browser, network and long-term memory. Agents stay online around the clock, even when the user's laptop is closed.

You can use your own API keys with Memoh's built-in agent, or host existing Claude Code and Codex agents inside Memoh workspaces. Users talk to agents through Telegram, Discord, Lark, WeChat or a Web UI. Agents keep context across sessions and platforms, drive a browser, call MCP tools and run scheduled tasks.

The full stack can be self-hosted with Docker Compose, or accessed through the hosted Memoh Cloud service.

Key features

  • 01Each agent gets an isolated workspace with its own filesystem, desktop, browser and network.
  • 02Long-term memory persists context across sessions and platforms.
  • 03Connects to Telegram, Discord, Lark, WeChat and a Web UI.
  • 04Hosts existing Claude Code and Codex agents inside workspaces.
  • 05Calls MCP tools and runs scheduled tasks.
  • 06Supports self-hosting with Docker Compose.

Use cases

  • Running a personal AI assistant that stays online and remembers past conversations.
  • Assigning a dedicated agent to each member of a family or team.
  • Hosting coding agents such as Claude Code or Codex in a shared environment.
  • Automating browser tasks and scheduled jobs through chat apps.

Quick start

  1. 1

    Clone the repository with submodules

    git clone --depth 1 --recurse-submodules --shallow-submodules https://github.com/felinics/Memoh.git
  2. 2

    Enter the project directory

    cd Memoh
  3. 3

    Copy the Docker config template to config.toml and edit it

    cp conf/app.docker.toml config.toml
  4. 4

    Set MEMOH_INTERNAL_RPC_SHARED_SECRET, then start the stack

    docker compose up -d

Things to know

  • Self-hosting requires Docker, a config.toml edit, and an internal RPC secret.
  • The built-in agent needs your own API keys from a model provider.
  • Licensed under AGPL-3.0, a strong copyleft license that applies to networked modifications.

Topics

FAQ

What is Memoh?
Open-source multi-agent platform that gives each AI agent its own cloud workspace with filesystem, desktop, browser, network and long-term memory. Memoh runs AI agents in isolated cloud workspaces, so each agent has its own filesystem, desktop, browser, network and long-term memory. Agents stay online around the clock, even when the user's laptop is closed.
Is Memoh free to use? What license does it have?
Memoh is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
Is Memoh still maintained?
Actively maintained. The last commit to Memoh was on Oct 22, 2026.
What can I use Memoh for?
Running a personal AI assistant that stays online and remembers past conversations. Assigning a dedicated agent to each member of a family or team. Hosting coding agents such as Claude Code or Codex in a shared environment. Automating browser tasks and scheduled jobs through chat apps.
How do I install Memoh?
Clone the repository with submodules: git clone --depth 1 --recurse-submodules --shallow-submodules https://github.com/felinics/Memoh.git Enter the project directory: cd Memoh Copy the Docker config template to config.toml and edit it: cp conf/app.docker.toml config.toml Set MEMOH_INTERNAL_RPC_SHARED_SECRET, then start the stack: docker compose up -d
Where can I get Memoh?
The source code and installation instructions are on GitHub: https://github.com/felinics/Memoh

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects