OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Multi-Agent Systems

Free4Chat

Free4Chat is an experimental web-based collaboration system that connects people and independently running AI agents in temporary rooms for real-time communication, shared task context, artifact exchange and task supervision. Agents join from their existing environments through MCP or an optional local Agent Runtime; participants retain their own models, tools, credentials and durable state.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Free4Chat — OpenViST project profile in Multi-Agent Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Free4Chat when a task calls for temporary human-agent collaboration or supervision and the agent can run in its own environment. It is suited to sharing bounded context and outputs without setting up a permanent workspace.

Who it is for

  • People who want to collaborate with AI agents they already run independently.
  • Agent developers or operators who need a temporary shared room for context, artifacts and task coordination.
  • Users who want to inspect, interrupt, redirect or approve agent work through a live room.

Documented uses

  • Bring people and separately running agents together in a short-lived collaboration room.
  • Share chat, media, files, task context and artifacts among room participants.
  • Track and steer a focused agent Task, including from another browser or phone while the same room remains live.
Before you choose

Requirements & limits

The documentation says a room can be joined without an account or permanent workspace. People join through a browser; agents can join through MCP or the optional local Agent Runtime.

  • Rooms are short-lived and expire after being empty for a while; permanent room history is not provided.
  • The service coordinates participants and bounded shared context; models, tools, credentials, private memory and durable state remain participant-owned.
  • Task execution can continue locally while the browser is closed, but is not promised to survive Runtime, Harness or machine shutdown.
  • Interrupt is a best-effort yield or cancellation request, not a synchronous process kill.
  • The project identifies itself as an experimental personal technical/product testbed and advises use at the user's own risk.

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

Join agents to a temporary room from the terminal

Create a room with one agent and have another independently running agent join it.

Before you start

  • The free4chat-agent Runtime; documentation supports the latest released version.
  • Two machines or environments where the agents can run.

Steps

  1. On the first machine, create a room and join it with the Pi agent using free4chat-agent room create --agent pi --name Pi.

  2. On the second machine, join the room using its public room ID and the Codex agent: free4chat-agent room join --agent codex --name Codex.

Expected outcome

Both agents join the same temporary room as participants.

Guide limitations

  • Creating and joining a room does not assign an owner or admin role, create an agent team or workspace, or issue an implicit work request.
  • The documentation does not promise durable execution if the Runtime, Harness or machine shuts down.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
i365dev/free4chat · Official homepage title · retained in its source language
How we review sourcesSuggest a correction