FLUJO

MIT License v3.48.3 Local-first Open Source

An open-source, local-first platform for building MCP-powered AI workflows — model management, MCP servers, a visual flow builder, chat, and headless automation in one app. Workspace data is stored locally. Cloud AI and connected apps receive the requests you choose to send. Anonymous usage telemetry is optional in Settings.

▶ FLUJO — Your AI. In Flow.

Get started in one command

PS> irm https://raw.githubusercontent.com/mario-andreschak/FLUJO/main/scripts/install.ps1 | iex

Press Start, type powershell, press Enter, then paste the command. Installs to %LOCALAPPDATA%\FLUJO by default; an uninstaller is included.

$ curl -fsSL https://raw.githubusercontent.com/mario-andreschak/FLUJO/main/scripts/install.sh | bash

Installs prerequisites via your package manager (Homebrew on macOS), clones and builds FLUJO into ~/FLUJO, and registers the flujo command.

$ npx flujo-ai@latest

Fastest start with Node.js 22.19+ within 22.x or 24.2+ within 24.x — runs the latest published, prebuilt FLUJO with no clone or build. Data lives in ~/.flujo; some MCP servers may still need git / python / uv on your PATH.

$ git clone https://github.com/mario-andreschak/FLUJO.git && cd FLUJO && docker compose up --build

Your flows, keys, and MCP servers persist in named volumes across restarts. The default local profile binds to localhost only. For deliberate network access, follow the network setup guide and use an authenticating reverse proxy. To update later, git pull then re-run the same command — --build rebuilds the image with the new code (plain docker compose up would reuse the cached image and silently run the old version).

The Windows / Linux / macOS installers also set up everything FLUJO needs — Git, Node.js, Python & uv. Then open http://localhost:4200. Read the setup guide for installation choices, updates, data locations, and troubleshooting.

Key Features

From API-key vault to headless automation — the full path from idea to running agent.

Secure Key Vault

API keys are encrypted at rest and never sent to the browser — bind a key once as a global variable and reuse it in any model or MCP server config.

Model Management

OpenAI, Anthropic, Google Gemini, X.ai, OpenRouter, and local Ollama — plus using your Claude Pro/Max subscription directly instead of a metered API key.

MCP Marketplace

Search the official MCP Registry and install servers with one click, pick from the curated Spotlight, or install from GitHub / a local folder. Inspect and test every tool in-app.

Visual Flow Builder

Drag-and-drop orchestration with branching, handoffs, loops, and subflows — reuse a whole flow like a function, with per-node tool and prompt scoping.

Chat, Live View & Debugger

Watch runs progress node-by-node with token usage, set breakpoints and step through executions, and require human approval before any tool call.

Planned Executions

Run flows headlessly on cron schedules, webhooks, file changes, MCP tool polling, or URL watches — with full run history per trigger.

OpenAI-compatible Endpoint

Point Cline, Roo Code, Cursor, or any OpenAI-SDK client at FLUJO and call your flows as models — use FLUJO as the backend for your other AI tools.

MCP Proxy

Configure an MCP server once in FLUJO, then re-expose it to Claude Desktop, Cursor, Cline, and any other MCP client over Streamable HTTP.

Built-in API Docs

A searchable, curated in-app HTTP reference for chat, flows, models, MCP, automations, and administration, with a source-derived route inventory in the repository.

The workshop website built and reviewed with a FLUJO Website Builder agent

Build a website. Keep the builder.

Connect your AI and file tools, save a reusable Website Builder agent, then give it a brief. Open the website, review it, and ask for improvements in the same chat.

Follow the Website Builder tutorial →

Plug it into your tools

FLUJO speaks the protocols your other tools already use.

Flows as OpenAI-compatible models

Point Cline, Roo Code, Cursor, or any OpenAI-SDK client at FLUJO and select your flow as the model — use the credentials required by your deployment. Network access setup.

Base URL:  http://localhost:4200/v1
Model:     flow-<your-flow-name>

Configured MCP servers, re-exposed

Set a server up once in FLUJO — env vars, auth, build — then use it from Claude Desktop, Cursor, or Cline via the proxy endpoint.

http://localhost:4200/mcp-proxy/<server-name>

FLUJO

Everything orbits your flows