OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Flujo

Flujo is an open-source, local-first TypeScript framework for visually building and running AI agents as recipes or graph-based workflows, inspecting their runs, connecting multiple AI models and applications, and exposing completed agents through MCP and OpenAI-compatible endpoints.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Flujo — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Flujo when an agent workflow needs visual composition, multiple model or application connections, run inspection, and a path to reuse the result through MCP or an OpenAI-compatible interface.

Who it is for

  • Developers building internal AI-agent automation.
  • Teams needing a private workspace for multi-model workflows.
  • Builders integrating MCP-enabled applications with reusable agents.
  • Operators who want visual workflow inspection and debugging.

Documented uses

  • Create step-by-step or graph-based multi-agent workflows.
  • Connect AI providers, local models and application capabilities through MCP.
  • Inspect and debug workflow runs.
  • Expose built agents to other software through MCP or OpenAI-compatible endpoints.
  • Keep model credentials and workspace data in a local-first environment.
Before you choose

Requirements & limits

The project is open source under the MIT license. The repository documents self-hosted installation on Windows, Linux and macOS, plus a prebuilt npm route. It also links to an online Flujo instance, but the supplied material does not establish its account, pricing or availability terms.

  • Choose the visual builder when workflows need recipe-style steps or connected graph nodes.
  • Use the MCP features when applications or tools need to be configured, inspected or re-exposed through Flujo.
  • Consider the local-first model when keeping secrets server-side and retaining workspace control matters.
  • The documented installer includes Git, Node.js, Python, uv and ripgrep; the npm route may still need Git, Python and uv for MCP servers.
  • MCP client compatibility is evidenced through the MCP proxy description, but the specific supported client set beyond the named examples is not specified.

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 Flujo with the recommended installer

Set up the Flujo workspace and its documented runtime dependencies.

Before you start

  • Windows, Linux or macOS
  • Internet access for the installer and project download

Steps

  1. On Windows, download and run the latest flujo-setup.exe, or execute the documented PowerShell installer.

  2. On Linux or macOS, run the documented install.sh command in a terminal.

  3. The installer sets up Git, Node.js, Python, uv and ripgrep, clones Flujo, builds it and creates a global flujo command.

  4. If Node.js is already installed, use the documented npx flujo-ai route instead; MCP servers may still require Git, Python and uv on PATH.

Expected outcome

A locally installed Flujo workspace with the documented command-line entry point, or a prebuilt npm launch path.

Guide limitations

  • The source describes installation stages but does not provide a verification result for a particular machine.
  • The npx route does not itself supply all runtimes that MCP servers may require.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
4 October 2026
Original website title
mario-andreschak/FLUJO · Official homepage title · retained in its source language
How we review sourcesSuggest a correction