OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

IM.codes

IM.codes is an open-source coordination service for AI coding agents, combining shared memory and managed MCP tools with session supervision, cross-provider audits and controlled computer access.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

IM.codes — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use IM.codes when coding work spans agent providers or machines and you need shared context, session visibility, supervision or controlled computer access in the same workflow.

Who it is for

  • Developers and teams coordinating coding work across multiple AI agents or providers.
  • Operators who want to supervise agent sessions and expose enrolled computers for scoped agent actions.
  • People who want to self-host a shared service for coding-agent workflows.

Documented uses

  • Carry reusable context from completed work between supported coding agents.
  • Review terminal sessions, files, Git changes and localhost previews from a connected interface.
  • Supervise multi-agent work and conduct cross-provider audits.
  • Let authorized agents run scoped commands, transfer files or control applications on enrolled computers.
  • View or take control of capable Windows nodes through a browser.
Before you choose

Requirements & limits

The project describes app.im.codes as a testing instance and recommends self-hosting for anything beyond evaluation. Its shared instance has no uptime guarantees and may be rate-limited. The repository is public and MIT-licensed.

  • Self-hosting requires Docker and a domain with DNS pointed at the server.
  • The daemon requirements include Node.js 22 or later, an AI coding agent, and tmux on macOS or Linux.
  • Windows support is labeled experimental and uses built-in ConPTY.
  • Controlled Nodes have separate credentials and owner-revocable execution access.
  • The project provides no warranty, SLA, or guarantees of stability, security or backward compatibility.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a self-hosted IM.codes server and daemon

Deploy the server and bind the local machine as a daemon using the documented one-command setup.

Before you start

  • Docker
  • A domain with DNS pointing to the server
  • Node.js 22 or later
  • An AI coding agent
  • tmux on macOS or Linux; Windows uses ConPTY and is marked experimental

Steps

  1. Install the IM.codes CLI globally with npm.

  2. Create and enter an installation directory.

  3. Run the setup command with the domain pointed at the server.

  4. Use the generated credentials to sign in; the setup starts PostgreSQL, the server and Caddy, then binds the local daemon.

Expected outcome

A self-hosted server and local daemon are configured, with credentials printed at the end of setup.

Guide limitations

  • The setup documentation requires Docker and a domain with DNS pointing to the server.
  • The project states that it has no warranty, SLA, or guarantees of stability, security or backward compatibility.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
im4codes/imcodes · Official homepage title · retained in its source language
How we review sourcesSuggest a correction