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

Openflow

OpenFlow is a desktop visual builder for multi-agent AI workflows. Users arrange role cards on a canvas, connect them into pipelines, save workflows, and run connected agents in parallel. A fork of OpenCode’s headless engine drives the agents; OpenFlow’s own code is maintained in the separate packages/flow workspace. The project is TypeScript and MIT-licensed.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Openflow — 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 OpenFlow when a task benefits from coordinating multiple role-based agents in a connected, visual pipeline, especially when agents need to work against a project directory.

Who it is for

  • Developers coordinating several AI agents in a role-based pipeline.
  • OpenCode users seeking a visual canvas for designing and running parallel agent workflows.
  • People who want to save multi-agent workflow arrangements for reuse.

Documented uses

  • Arrange roles such as planner, architect, and coder as workflow cards.
  • Connect agent roles into a pipeline and save the workflow.
  • Run connected agents in parallel against a project directory.
  • Start the local engine and canvas from the source repository.
Before you choose

Requirements & limits

  • The documented setup requires Bun 1.3 or newer and Git.
  • The interface and engine use ports 5174 and 4096 respectively.
  • Agents can read and edit real files in the selected project directory.
  • The OpenCode engine is a fork and drives the agents; OpenFlow’s own code is in packages/flow.
  • The README identifies MIT as the project license.

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

Install and start OpenFlow locally

Set up the repository and launch its engine and workflow canvas together.

Before you start

  • Bun 1.3 or newer
  • Git

Steps

  1. Clone https://github.com/SeeRay11/OpenFlow.git and change into the OpenFlow directory.

  2. Run bun install to install the workspace, including the OpenCode engine and OpenFlow code.

  3. Start the processes with FLOW_MANAGE_SERVER=1 bun openflow.ts on bash, zsh, Git Bash, WSL, macOS, or Linux; use the README’s equivalent environment-variable command for another shell.

  4. Open the locally served canvas at http://localhost:5174; the engine uses port 4096.

Expected outcome

The local engine and canvas start together, with the canvas available at localhost:5174.

Guide limitations

  • The initial install is large because it downloads engine dependencies.
  • The selected project is read and written by agents, which can edit real files.
  • Another engine process may trigger a shared-database lock; freeing ports alone does not resolve that condition.
  • This is source-based setup guidance, not a human execution or compatibility test.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
SeeRay11/OpenFlow · Official homepage title · retained in its source language
How we review sourcesSuggest a correction