An n8n alternative where a team does the work

Keep your n8n workflows. Your Murage bots run them as tools, and a workflow can start a Murage task. Murage adds a Chief of Staff and a team that decides when to run them, on your computer.

The short answer

Keep the workflows you trust. Your Murage team calls them when the job needs one.

Which should you choose?

Murage

Open source
On your computer
The AI you choose

Who does the work: A Chief of Staff and a team

n8n

Free to self-host
Their cloud or your server
Workflows you build

Who does the work: Workflows and agents you build

Download Murage

Checked October 2026 (October 9, 2026). n8n’s own pages and docs for everything about n8n.

Keep your workflows. Let your bots run them.

You talk to one Chief of Staff. It builds the team, and each bot runs the engine that suits its job, on the plan or key you already have. An engine is the AI that does a bot’s thinking. To run teams and use tools, pick a tool-capable one: the built-in Fuigo engine with a supported model, Claude Code, Codex or Hermes. A plain OpenAI-compatible endpoint works for chat and drafting only. Portraits made with Murage’s own avatar generator. See every engine

Keep your workflows. Your bots run them.

Your n8n workflows become tools your team can use.

Your bots run your n8n workflows as tools through n8n’s MCP server (MCP servers are plug-ins that give bots new tools), and a workflow can start a Murage task through a local webhook (cloud senders need a tunnel). Deeper built-in n8n support is coming. n8n keeps doing what it does best: workflows you can see and control, including AI agent steps. Your Chief of Staff decides when a job needs one.

What Murage adds on top

A Chief of Staff that takes the brief and decides which bot, engine or workflow each step needs
Bots that research, write and build in your folders, then hand the fixed steps to your workflows
1,600+ apps through Flux Router credits or your own connected-apps key, next to the ones your workflows already reach
Memory, files and chats stored on your computer

Why Murage is open source Every engine Murage runs

At a glance.

Ten things buyers weigh up, including the ones where they’re ahead.

What matters

Murage

n8n

The app needs no extra subscription
It runs on the AI plan or key you already have
Yes
Open source desktop app; Murage Cloud is optional and paid
Source
Yes
Community edition, free to self-host
Source
You choose which AI company does the work
Per bot or per task, from different vendors
Yes
Per bot: Fuigo, Claude Code, Codex, Hermes, a local model and more
Yes
Cloud or offline models, swappable without rebuilding a workflow
Source
Named agents that keep their memory
A standing team that lasts beyond one task
Yes
Named bots, each with its own memory, kept on your computer
Partly
Multi-agent setups you build on the canvas
Source
Asks before acting outside, by default
Messaging someone new, posting, paying, deleting
Yes
New bots start on Ask; on Full access the stop line judges what each engine asks
Source
Partly
A human review step you add to an AI Agent node; the approver sees the tool and its parameters
Source
Memory stored on your computer
Not in a vendor’s cloud
Yes
Memory, files and chats on your computer
Partly
On your server if you self-host; otherwise n8n’s cloud
Source
Open source
A license you can read
Yes
AGPL-3.0
Source
Partly
Source available under its Sustainable Use License
Source
Reach it from your phone
Briefs and approvals away from the desk
Yes
An iPhone and Android app, plus your Chief in Telegram, Slack, Discord or WhatsApp
Not stated
Not stated
Source
Keeps working with your computer off, today
Hosted or on a server
Yes
On any computer you leave on, including your own remote server. Murage Cloud (coming soon) will run one for you
Yes
In n8n’s cloud or on your server
Source
Native phone app
An iOS or Android app, not a web app or chat bot
Yes
iPhone and Android
Not stated
Not stated
Source
SOC 2 report
What a procurement team will ask for
No
No SOC 2 report
Yes
A SOC 2 report, available to enterprise customers
Source

Each of n8n’s marks links the page it comes from, checked October 9, 2026. “Not stated” means we couldn’t find it on the page named, not that it’s missing.

Both apps are open source. AI usage and any hosting or optional services are separate. n8n is free to self-host on a server you run. n8n Cloud starts at €20 a month billed annually (Starter, 2,500 executions).

See one job from brief to result

Where your work lives.

Murage

Your computer

MemoryFilesChats

Your memory, files and chats are stored on your computer. What a bot sends to your AI, and to any app or service you connect, goes to them.

n8n

A machine you host

MemoryConfig

n8n runs in its cloud, or on a server you set up and keep running.

Who does the work.

You brief one Chief of Staff. It builds a team of specialists for the job, and the team keeps its memory for next time. You can put a reviewer on any team to check work before it reaches you. Portraits made with Murage’s own avatar generator.

n8n

You build each workflow or agent on the canvas, step by step. Your Murage bots can run those workflows as tools.

What Murage adds for an n8n user.

The brief

“Go through my inbox before I get in. Draft replies in my voice. Send nothing.”

What came back

Twelve emails came in. The team drafted eight replies, set four aside, and left two questions only the owner could answer. Nothing was sent.

Needs from you, in the team’s own report

Dana Brooks, Acme Studio · RenewalThe prior contract terms
Jordan Lee, Brightline Design · Pricing for 20 seatsThe per-seat price, and whether we offer a pilot
Priya S. · Export to PDF?Confirm the exact steps
Ben C. · Refund requestMoney. Check the billing record first

Replay of a real run on demo data. The company is made up. Engine: Flux Auto, worked 1m 41s. Watch the run

Keep your workflows. Add a team on top

You don’t have to choose. Switch on n8n’s MCP server in Murage and your bots run your workflows as tools, or let a workflow start a Murage task with a webhook.

Describe the outcome, not the steps

Brief your Chief of Staff in plain English, and it builds a team of named bots, each with a role, memory and the apps it needs. Keep the canvas for the steps that must never change.

Bots that do the work itself

Each bot runs Fuigo, Claude Code, Codex, Hermes, a local model or an API key, so it can research, write and build in your folders, then hand fixed steps to a workflow.

Asks first, by default

New bots start on Ask. On Full access, a bot stops before messaging anyone new, posting publicly, paying, or deleting outside its folder, unless you have allowed it for that task. On Hermes and other ACP engines, Ask and the stop line cover only what the engine asks Murage about.

Where n8n is ahead.

Fair’s fair. If one of these matters most to you, it may be the better choice today.

1
Always on, with triggers
Workflows fire on schedules and app events around the clock, in n8n’s cloud or on your server. Murage runs on whichever computer it is installed on, so for work around the clock you put it on a machine you leave on, such as your own remote server. Murage Cloud (coming soon) will run one for you.
2
Its integrations, for your bots too
Its 1,600+ integrations are there for your bots too, through your workflows.
3
Repeatable, inspectable workflows
You can inspect every execution on the canvas. That suits a process that must run the same way every time.
4
Self-hosting and compliance
Run the Community edition on your own server for free. n8n aligns its security program to SOC 2, with annual independent audits.

Which should you choose?

Keep the workflows you trust. Your Murage team calls them when the job needs one.

Choose Murage if

You’d rather describe the job to a Chief of Staff than wire every step
You want bots that research, write and build in your folders, then hand fixed steps to your workflows
You want memory and files stored on your own computer
You want everyone on your staff to have it, with no Murage seat fees

Download Murage

Choose n8n if

You need triggers firing around the clock today, in n8n’s cloud or on your server
You need a process that runs the same way every time, with every execution inspectable
You want a visual canvas you control step by step (you can keep it: Murage bots can run your n8n workflows through MCP)

Still deciding? Read how we check that a job is finished, then try one of your own.

Every detail.

Getting in

Price to start
MurageOpen source desktop app; Murage Cloud is an optional paid hosted service
n8nCommunity edition is free to self-host. n8n Cloud: Starter €20 a month billed annually (2,500 executions), Pro €50 (10,000), Business €667 (40,000, self-hosted)
How you start
MurageDownload Murage, pick an engine, meet Ember and describe the job
n8nSign up for n8n Cloud, or self-host with Docker Compose or the one-line setup
How you set it up
MurageDescribe the job to Ember, your Chief of Staff; it builds the team
n8nA visual canvas of nodes, with JavaScript or Python where you want code
What usage costs
MurageNothing from us. Your AI provider bills as usual. Flux Router, Ferrox Labs’ optional credit-billed service, covers AI models, connected apps, voice, images and web search if you use it; your own keys work instead
n8nn8n Cloud plans include a monthly number of workflow executions; the Community edition runs on your own server

Where it runs

Runs on
MurageOn your computer
n8nn8n Cloud (hosted by n8n on Azure in the EU, currently), or your own server
Keeps working with your laptop closed
MurageYes, on any computer you leave on, including your own remote server. Murage Cloud (coming soon) will look after one for you
n8nYes, in n8n Cloud or on your server
Air-gapped or on-premises
MurageRuns on your computer; the AI engine you pick decides what leaves it, and local models stay on the machine
n8nYes: n8n says you can deploy on-premises, air-gapped if you need it
Platforms
MurageMac, Windows and Linux, plus an iPhone app and an Android app; Telegram on your phone
n8nA web app; self-host anywhere Docker runs

Brains

Engines
MurageThe engine you choose for each bot: Fuigo (built in), Claude Code, Codex, Hermes, Grok Build, OpenCode, local models or your own API keys
n8nConnect OpenAI, Anthropic, Google or open-source models, cloud or offline
Bring the AI plan you already pay for
MurageYes: Claude, Codex or Grok plans, local models or your own keys
n8nYou add your own model keys; Starter and above also list AI models without API keys
Swap the AI behind the work
MurageYes, per bot, at any time; memory, skills and approvals stay put
n8nYes: n8n says you can switch providers without changing your workflows

The team

Who leads
MurageEmber, your Chief of Staff, builds the team for each job and hands out the work
n8nYou do: you design each workflow and agent on the canvas
Multiple agents
MurageA Chief of Staff builds and runs a team for each job
n8nMulti-agent setups you build and wire yourself
Projects with boards
MurageYes: projects split a job into cards on a board, with a lead, a plan, files and sign-off
n8nProjects in n8n group workflows and credentials for sharing (1 on Starter, 3 on Pro, 6 on Business); the pricing page describes no task board
Routines and schedules
MurageRoutines run a bot task or a team goal on a schedule while your computer is on; run now, pause and resume any time
n8nSchedule Trigger and Webhook nodes start workflows around the clock

Hands

Apps
Murage1,600+ apps through Flux Router or your own connected-apps key, plus any public API and any MCP, and it builds the connector itself if one doesn’t exist
n8n1,600+ integrations per n8n’s GitHub README; n8n.io’s homepage says over 500
Any public API or MCP
MurageYes: a bot works out the connection itself, and you can paste an MCP server link
n8nCustom API connections and HTTP requests you configure; n8n says you can connect to anything that speaks MCP
Run code
MurageBots on Claude Code or Codex write and run code in your folders
n8nA Code node for JavaScript, Python and npm packages inside a workflow
Your own signed-in browser
MurageMurage for Chrome works in your own signed-in Chrome, in the tabs you share
n8nNot described on n8n’s homepage
Phone
MurageiPhone and Android app that opens your workspace: approvals, questions and calls with your team
n8nNot described on n8n’s homepage

Control

Asks before acting
MurageNew bots start on Ask. On Full access, a bot stops before messaging anyone new, posting publicly, paying, or deleting outside its folder, unless you have allowed it for that task. Replies in an existing conversation go ahead on Auto and Full access. No limits turns the stop line off for your own chats; routines still stop at it. On Claude Code and Codex, the stop line also covers tools from MCP servers you add. On Hermes and other ACP engines, Ask and the stop line cover only what the engine asks Murage about.
n8nA human review step you add to an AI Agent node: the approver sees the tool and its parameters and approves or denies
Where approvals arrive
MurageIn the app and in Telegram, Slack, Discord or WhatsApp
n8nn8n’s chat, Slack, Discord, Telegram, Teams, Gmail, WhatsApp Business, Google Chat or Outlook
Lessons you can see, edit and undo
MurageYes: corrections become lessons you can read, change or take back
n8nNot described on n8n’s homepage; you edit the workflow itself
Inspecting a run
MurageChats, files and bot activity are on your computer, and each routine run leaves a record
n8nEvery step of a run is traceable on the canvas; Cloud keeps execution logs for 7 days on Starter and 30 on Pro
Open source
MurageYes, AGPL-3.0
n8nSource available under the Sustainable Use License (fair-code)

Your data

Where memory lives
MurageDesktop: on your computer, open to read and edit. Cloud: on your EU-hosted server, encrypted with a key Ferrox Labs holds
n8nOn n8n Cloud, or on your server if you self-host
Compliance
MurageYour files and memory stay on your computer, and the code is open to read
n8nn8n aligns its security program to SOC 2 with annual independent audits; the SOC 2 report goes to enterprise customers
How it signs in
MurageYou sign in to your AI provider yourself; Murage doesn’t handle your Claude login
n8nCredentials live in n8n; Enterprise adds external secret stores

Reach

Chat apps
MurageMessage your Chief of Staff on Telegram, Slack, Discord or WhatsApp
n8nChat and messaging nodes inside workflows, plus n8n’s built-in chat
Voice
MurageCall any bot in the app and just talk; bots leave voice notes. Ask your Chief for a voice note in Telegram, Slack, Discord or WhatsApp, and on WhatsApp send it one of your own
n8nNot described on n8n’s homepage
Working together
MurageBots run your n8n workflows through n8n’s MCP server, and a workflow can start a Murage task by webhook
n8nn8n’s instance-level MCP server exposes the workflows you switch on
For your staff
MurageEach person runs Murage on their own computer; export a team and everyone imports it
n8nUnlimited users on every n8n Cloud plan

Question	Murage	n8n
Getting in
Price to start	Open source desktop app; Murage Cloud is an optional paid hosted service	Community edition is free to self-host. n8n Cloud: Starter €20 a month billed annually (2,500 executions), Pro €50 (10,000), Business €667 (40,000, self-hosted)
How you start	Download Murage, pick an engine, meet Ember and describe the job	Sign up for n8n Cloud, or self-host with Docker Compose or the one-line setup
How you set it up	Describe the job to Ember, your Chief of Staff; it builds the team	A visual canvas of nodes, with JavaScript or Python where you want code
What usage costs	Nothing from us. Your AI provider bills as usual. Flux Router, Ferrox Labs’ optional credit-billed service, covers AI models, connected apps, voice, images and web search if you use it; your own keys work instead	n8n Cloud plans include a monthly number of workflow executions; the Community edition runs on your own server
Where it runs
Runs on	On your computer	n8n Cloud (hosted by n8n on Azure in the EU, currently), or your own server
Keeps working with your laptop closed	Yes, on any computer you leave on, including your own remote server. Murage Cloud (coming soon) will look after one for you	Yes, in n8n Cloud or on your server
Air-gapped or on-premises	Runs on your computer; the AI engine you pick decides what leaves it, and local models stay on the machine	Yes: n8n says you can deploy on-premises, air-gapped if you need it
Platforms	Mac, Windows and Linux, plus an iPhone app and an Android app; Telegram on your phone	A web app; self-host anywhere Docker runs
Brains
Engines	The engine you choose for each bot: Fuigo (built in), Claude Code, Codex, Hermes, Grok Build, OpenCode, local models or your own API keys	Connect OpenAI, Anthropic, Google or open-source models, cloud or offline
Bring the AI plan you already pay for	Yes: Claude, Codex or Grok plans, local models or your own keys	You add your own model keys; Starter and above also list AI models without API keys
Swap the AI behind the work	Yes, per bot, at any time; memory, skills and approvals stay put	Yes: n8n says you can switch providers without changing your workflows
The team
Who leads	Ember, your Chief of Staff, builds the team for each job and hands out the work	You do: you design each workflow and agent on the canvas
Multiple agents	A Chief of Staff builds and runs a team for each job	Multi-agent setups you build and wire yourself
Projects with boards	Yes: projects split a job into cards on a board, with a lead, a plan, files and sign-off	Projects in n8n group workflows and credentials for sharing (1 on Starter, 3 on Pro, 6 on Business); the pricing page describes no task board
Routines and schedules	Routines run a bot task or a team goal on a schedule while your computer is on; run now, pause and resume any time	Schedule Trigger and Webhook nodes start workflows around the clock
Hands
Apps	1,600+ apps through Flux Router or your own connected-apps key, plus any public API and any MCP, and it builds the connector itself if one doesn’t exist	1,600+ integrations per n8n’s GitHub README; n8n.io’s homepage says over 500
Any public API or MCP	Yes: a bot works out the connection itself, and you can paste an MCP server link	Custom API connections and HTTP requests you configure; n8n says you can connect to anything that speaks MCP
Run code	Bots on Claude Code or Codex write and run code in your folders	A Code node for JavaScript, Python and npm packages inside a workflow
Your own signed-in browser	Murage for Chrome works in your own signed-in Chrome, in the tabs you share	Not described on n8n’s homepage
Phone	iPhone and Android app that opens your workspace: approvals, questions and calls with your team	Not described on n8n’s homepage
Control
Asks before acting	New bots start on Ask. On Full access, a bot stops before messaging anyone new, posting publicly, paying, or deleting outside its folder, unless you have allowed it for that task. Replies in an existing conversation go ahead on Auto and Full access. No limits turns the stop line off for your own chats; routines still stop at it. On Claude Code and Codex, the stop line also covers tools from MCP servers you add. On Hermes and other ACP engines, Ask and the stop line cover only what the engine asks Murage about.	A human review step you add to an AI Agent node: the approver sees the tool and its parameters and approves or denies
Where approvals arrive	In the app and in Telegram, Slack, Discord or WhatsApp	n8n’s chat, Slack, Discord, Telegram, Teams, Gmail, WhatsApp Business, Google Chat or Outlook
Lessons you can see, edit and undo	Yes: corrections become lessons you can read, change or take back	Not described on n8n’s homepage; you edit the workflow itself
Inspecting a run	Chats, files and bot activity are on your computer, and each routine run leaves a record	Every step of a run is traceable on the canvas; Cloud keeps execution logs for 7 days on Starter and 30 on Pro
Open source	Yes, AGPL-3.0	Source available under the Sustainable Use License (fair-code)
Your data
Where memory lives	Desktop: on your computer, open to read and edit. Cloud: on your EU-hosted server, encrypted with a key Ferrox Labs holds	On n8n Cloud, or on your server if you self-host
Compliance	Your files and memory stay on your computer, and the code is open to read	n8n aligns its security program to SOC 2 with annual independent audits; the SOC 2 report goes to enterprise customers
How it signs in	You sign in to your AI provider yourself; Murage doesn’t handle your Claude login	Credentials live in n8n; Enterprise adds external secret stores
Reach
Chat apps	Message your Chief of Staff on Telegram, Slack, Discord or WhatsApp	Chat and messaging nodes inside workflows, plus n8n’s built-in chat
Voice	Call any bot in the app and just talk; bots leave voice notes. Ask your Chief for a voice note in Telegram, Slack, Discord or WhatsApp, and on WhatsApp send it one of your own	Not described on n8n’s homepage
Working together	Bots run your n8n workflows through n8n’s MCP server, and a workflow can start a Murage task by webhook	n8n’s instance-level MCP server exposes the workflows you switch on
For your staff	Each person runs Murage on their own computer; export a team and everyone imports it	Unlimited users on every n8n Cloud plan

Same job, both ways.

Five real jobs, walked through with Murage and with n8n.

1
Send a weekly numbers report
With Murage
Brief Ember once: "every Monday at 9, pull last week’s numbers and write me a summary." It proposes a routine, you confirm it, and a bot runs it while your computer is on. Correct the wording once and it becomes a lesson.
With n8n
You build a workflow: a Schedule Trigger, a node per data source, a node to combine the numbers, an AI step to write the summary and a Gmail node to send it. It then runs the same way every time, in the cloud or on your server.
2
Triage the support inbox
With Murage
Ember sends an inbox bot through your connected mail. It sorts what matters and drafts replies, and it asks before anything is sent.
With n8n
You build a workflow with an email trigger and an AI Agent node, then add a human review step so a person approves each send.
3
Research three competitors and write a brief
With Murage
Ember builds a small team: a researcher per competitor and a writer. They share a project board, and the brief lands as a file in your folder.
With n8n
You wire search or HTTP nodes, a loop over the three companies, an agent to summarize each and a node that writes the document.
4
Sync invoices to your books every night
With Murage
This is n8n’s ground. Keep the workflow in n8n, switch it on for MCP, and let a Murage bot run it when a job needs it. Your Chief of Staff decides when.
With n8n
A scheduled workflow does the same steps every night, in n8n’s cloud or on your server, and you can inspect every execution.
5
Start a bot’s job from a workflow
With Murage
Create a local webhook in Routines, copy the address and secret, and point the workflow’s HTTP request step at it. Webhook-started tasks are judged as Auto, so risky actions wait for your approval.
With n8n
An HTTP Request node posts a JSON body with a task to Murage’s local address. If n8n runs on the same computer it can send straight to it; n8n Cloud needs a tunnel.

Already using n8n: your first day with Murage

You do not need to move anything. Keep n8n running, add Murage beside it, and hand over the jobs that need judgment first.

1
Install Murage on your Mac, Windows or Linux computer and pick an engine: the AI plan you already pay for, a local model or an API key.
2
Meet Ember, your Chief of Staff, and describe the job you most often rebuild by hand: research, drafting or follow-up.
3
Add n8n’s MCP server under Connected apps, then switch on in n8n the workflows you want bots to reach. n8n exposes nothing until you enable it.
4
Let the first run ask before it sends anything, and keep the fixed, high-volume workflows in n8n.
5
When a workflow should kick off a bot, create a local webhook in Routines and point an HTTP Request node at it.

Questions about Murage and n8n.

Does Murage work with n8n?

Yes, today, in both directions. Add n8n’s MCP server in Murage and switch it on (n8n exposes no workflow until you enable it in n8n), and every bot whose engine can use MCP tools can run your workflows. And an n8n workflow can start a Murage bot’s task by calling a Murage webhook. The webhook listens on your own computer, so a workflow on n8n Cloud needs a tunnel to reach it. Deeper, built-in n8n support is on the way.

Can Murage replace my n8n workflows?

For work that needs judgment, like research, drafting and follow-up, a Murage team is often simpler to set up. For fixed, high-volume automations that must fire around the clock, keep n8n. You can use both, and they already work together.

Does Murage connect to the same apps?

Murage connects 1,600+ apps, including Gmail, Slack, Notion and GitHub, through Flux Router credits or your own connected-apps key, plus MCP servers. n8n covers more, and your bots can reach those through your workflows.

Can my staff use it too?

Yes. Each person runs Murage on their own computer, with their own AI sign-in. Build a team once, export it, and everyone imports the same bots, skills and routines; logins never travel. Bots can post drafts to your team’s Slack channels through connected apps, with your approval. Shared workspaces for teams are coming soon to Murage Cloud.

Is Murage made by n8n?

No. Murage is an independent open-source project. It is not affiliated with or endorsed by n8n.

Sources, checked October 9, 2026

Murage is made by Ferrox Labs, a small team. Spotted something out of date? Tell us and we’ll fix it.

n8n is a trademark of its owner. Murage is not affiliated with or endorsed by n8n GmbH; the names and logos are used only to identify the products compared.

Get started in four steps

1
Download Murage
For Mac, Windows and Linux.
2
Pick an engine
Use the AI plan you already pay for, a local model or an API key.
3
Meet Ember
Your Chief of Staff, who builds the team for each job.
4
Give it the first job
Start small, and keep your n8n workflows for the steps that must never change. New bots ask before they act.

Download Murage All downloads and install notes

Give your first job to Murage.

Download the app, connect the AI you already pay for, and tell your Chief of Staff what needs doing. Plan on about ten minutes from install to a working team.

Download Murage

No account needed · Runs on the AI plan you already pay for

An n8n alternative where a team does the work

Keep your n8n workflows. Your Murage bots run them as tools, and a workflow can start a Murage task. Murage adds a Chief of Staff and a team that decides when to run them, on your computer.

The short answer

Keep the workflows you trust. Your Murage team calls them when the job needs one.

Which should you choose?

Murage

Open source
On your computer
The AI you choose

Who does the work: A Chief of Staff and a team

n8n

Free to self-host
Their cloud or your server
Workflows you build

Who does the work: Workflows and agents you build

Download Murage

Checked October 2026 (October 9, 2026). n8n’s own pages and docs for everything about n8n.

Question	Murage	n8n
Getting in
Price to start	Open source desktop app; Murage Cloud is an optional paid hosted service	Community edition is free to self-host. n8n Cloud: Starter €20 a month billed annually (2,500 executions), Pro €50 (10,000), Business €667 (40,000, self-hosted)
How you start	Download Murage, pick an engine, meet Ember and describe the job	Sign up for n8n Cloud, or self-host with Docker Compose or the one-line setup
How you set it up	Describe the job to Ember, your Chief of Staff; it builds the team	A visual canvas of nodes, with JavaScript or Python where you want code
What usage costs	Nothing from us. Your AI provider bills as usual. Flux Router, Ferrox Labs’ optional credit-billed service, covers AI models, connected apps, voice, images and web search if you use it; your own keys work instead	n8n Cloud plans include a monthly number of workflow executions; the Community edition runs on your own server
Where it runs
Runs on	On your computer	n8n Cloud (hosted by n8n on Azure in the EU, currently), or your own server
Keeps working with your laptop closed	Yes, on any computer you leave on, including your own remote server. Murage Cloud (coming soon) will look after one for you	Yes, in n8n Cloud or on your server
Air-gapped or on-premises	Runs on your computer; the AI engine you pick decides what leaves it, and local models stay on the machine	Yes: n8n says you can deploy on-premises, air-gapped if you need it
Platforms	Mac, Windows and Linux, plus an iPhone app and an Android app; Telegram on your phone	A web app; self-host anywhere Docker runs
Brains
Engines	The engine you choose for each bot: Fuigo (built in), Claude Code, Codex, Hermes, Grok Build, OpenCode, local models or your own API keys	Connect OpenAI, Anthropic, Google or open-source models, cloud or offline
Bring the AI plan you already pay for	Yes: Claude, Codex or Grok plans, local models or your own keys	You add your own model keys; Starter and above also list AI models without API keys
Swap the AI behind the work	Yes, per bot, at any time; memory, skills and approvals stay put	Yes: n8n says you can switch providers without changing your workflows
The team
Who leads	Ember, your Chief of Staff, builds the team for each job and hands out the work	You do: you design each workflow and agent on the canvas
Multiple agents	A Chief of Staff builds and runs a team for each job	Multi-agent setups you build and wire yourself
Projects with boards	Yes: projects split a job into cards on a board, with a lead, a plan, files and sign-off	Projects in n8n group workflows and credentials for sharing (1 on Starter, 3 on Pro, 6 on Business); the pricing page describes no task board
Routines and schedules	Routines run a bot task or a team goal on a schedule while your computer is on; run now, pause and resume any time	Schedule Trigger and Webhook nodes start workflows around the clock
Hands
Apps	1,600+ apps through Flux Router or your own connected-apps key, plus any public API and any MCP, and it builds the connector itself if one doesn’t exist	1,600+ integrations per n8n’s GitHub README; n8n.io’s homepage says over 500
Any public API or MCP	Yes: a bot works out the connection itself, and you can paste an MCP server link	Custom API connections and HTTP requests you configure; n8n says you can connect to anything that speaks MCP
Run code	Bots on Claude Code or Codex write and run code in your folders	A Code node for JavaScript, Python and npm packages inside a workflow
Your own signed-in browser	Murage for Chrome works in your own signed-in Chrome, in the tabs you share	Not described on n8n’s homepage
Phone	iPhone and Android app that opens your workspace: approvals, questions and calls with your team	Not described on n8n’s homepage
Control
Asks before acting	New bots start on Ask. On Full access, a bot stops before messaging anyone new, posting publicly, paying, or deleting outside its folder, unless you have allowed it for that task. Replies in an existing conversation go ahead on Auto and Full access. No limits turns the stop line off for your own chats; routines still stop at it. On Claude Code and Codex, the stop line also covers tools from MCP servers you add. On Hermes and other ACP engines, Ask and the stop line cover only what the engine asks Murage about.	A human review step you add to an AI Agent node: the approver sees the tool and its parameters and approves or denies
Where approvals arrive	In the app and in Telegram, Slack, Discord or WhatsApp	n8n’s chat, Slack, Discord, Telegram, Teams, Gmail, WhatsApp Business, Google Chat or Outlook
Lessons you can see, edit and undo	Yes: corrections become lessons you can read, change or take back	Not described on n8n’s homepage; you edit the workflow itself
Inspecting a run	Chats, files and bot activity are on your computer, and each routine run leaves a record	Every step of a run is traceable on the canvas; Cloud keeps execution logs for 7 days on Starter and 30 on Pro
Open source	Yes, AGPL-3.0	Source available under the Sustainable Use License (fair-code)
Your data
Where memory lives	Desktop: on your computer, open to read and edit. Cloud: on your EU-hosted server, encrypted with a key Ferrox Labs holds	On n8n Cloud, or on your server if you self-host
Compliance	Your files and memory stay on your computer, and the code is open to read	n8n aligns its security program to SOC 2 with annual independent audits; the SOC 2 report goes to enterprise customers
How it signs in	You sign in to your AI provider yourself; Murage doesn’t handle your Claude login	Credentials live in n8n; Enterprise adds external secret stores
Reach
Chat apps	Message your Chief of Staff on Telegram, Slack, Discord or WhatsApp	Chat and messaging nodes inside workflows, plus n8n’s built-in chat
Voice	Call any bot in the app and just talk; bots leave voice notes. Ask your Chief for a voice note in Telegram, Slack, Discord or WhatsApp, and on WhatsApp send it one of your own	Not described on n8n’s homepage
Working together	Bots run your n8n workflows through n8n’s MCP server, and a workflow can start a Murage task by webhook	n8n’s instance-level MCP server exposes the workflows you switch on
For your staff	Each person runs Murage on their own computer; export a team and everyone imports it	Unlimited users on every n8n Cloud plan

Murage

n8n

Getting in

Price to start

Open source desktop app; Murage Cloud is an optional paid hosted service

Community edition is free to self-host. n8n Cloud: Starter €20 a month billed annually (2,500 executions), Pro €50 (10,000), Business €667 (40,000, self-hosted)

How you start

Download Murage, pick an engine, meet Ember and describe the job

How you set it up

Describe the job to Ember, your Chief of Staff; it builds the team

A visual canvas of nodes, with JavaScript or Python where you want code

What usage costs

Nothing from us. Your AI provider bills as usual. Flux Router, Ferrox Labs’ optional credit-billed service, covers AI models, connected apps, voice, images and web search if you use it; your own keys work instead

n8n Cloud plans include a monthly number of workflow executions; the Community edition runs on your own server

Where it runs

Runs on

On your computer

n8n Cloud (hosted by n8n on Azure in the EU, currently), or your own server

Keeps working with your laptop closed

Yes, on any computer you leave on, including your own remote server. Murage Cloud (coming soon) will look after one for you

Yes, in n8n Cloud or on your server

Air-gapped or on-premises

Runs on your computer; the AI engine you pick decides what leaves it, and local models stay on the machine

Yes: n8n says you can deploy on-premises, air-gapped if you need it

Platforms

Mac, Windows and Linux, plus an iPhone app and an Android app; Telegram on your phone

A web app; self-host anywhere Docker runs

Brains

Engines

The engine you choose for each bot: Fuigo (built in), Claude Code, Codex, Hermes, Grok Build, OpenCode, local models or your own API keys

Connect OpenAI, Anthropic, Google or open-source models, cloud or offline

Bring the AI plan you already pay for

Yes: Claude, Codex or Grok plans, local models or your own keys

You add your own model keys; Starter and above also list AI models without API keys

Swap the AI behind the work

Yes, per bot, at any time; memory, skills and approvals stay put

Yes: n8n says you can switch providers without changing your workflows

The team

Who leads

Ember, your Chief of Staff, builds the team for each job and hands out the work

You do: you design each workflow and agent on the canvas

Multiple agents

A Chief of Staff builds and runs a team for each job

Multi-agent setups you build and wire yourself

Projects with boards

Yes: projects split a job into cards on a board, with a lead, a plan, files and sign-off

Projects in n8n group workflows and credentials for sharing (1 on Starter, 3 on Pro, 6 on Business); the pricing page describes no task board

Routines and schedules

Routines run a bot task or a team goal on a schedule while your computer is on; run now, pause and resume any time

Schedule Trigger and Webhook nodes start workflows around the clock

Hands

Apps

1,600+ apps through Flux Router or your own connected-apps key, plus any public API and any MCP, and it builds the connector itself if one doesn’t exist

1,600+ integrations per n8n’s GitHub README; n8n.io’s homepage says over 500

Any public API or MCP

Yes: a bot works out the connection itself, and you can paste an MCP server link

Custom API connections and HTTP requests you configure; n8n says you can connect to anything that speaks MCP

Run code

Bots on Claude Code or Codex write and run code in your folders

A Code node for JavaScript, Python and npm packages inside a workflow

Your own signed-in browser

Murage for Chrome works in your own signed-in Chrome, in the tabs you share

Not described on n8n’s homepage

Phone

iPhone and Android app that opens your workspace: approvals, questions and calls with your team

Not described on n8n’s homepage

Control

Asks before acting

New bots start on Ask. On Full access, a bot stops before messaging anyone new, posting publicly, paying, or deleting outside its folder, unless you have allowed it for that task. Replies in an existing conversation go ahead on Auto and Full access. No limits turns the stop line off for your own chats; routines still stop at it. On Claude Code and Codex, the stop line also covers tools from MCP servers you add. On Hermes and other ACP engines, Ask and the stop line cover only what the engine asks Murage about.

A human review step you add to an AI Agent node: the approver sees the tool and its parameters and approves or denies

Where approvals arrive

In the app and in Telegram, Slack, Discord or WhatsApp

n8n’s chat, Slack, Discord, Telegram, Teams, Gmail, WhatsApp Business, Google Chat or Outlook

Lessons you can see, edit and undo

Yes: corrections become lessons you can read, change or take back

Not described on n8n’s homepage; you edit the workflow itself

Inspecting a run

Chats, files and bot activity are on your computer, and each routine run leaves a record

Every step of a run is traceable on the canvas; Cloud keeps execution logs for 7 days on Starter and 30 on Pro

Open source

Yes, AGPL-3.0

Source available under the Sustainable Use License (fair-code)

Your data

Where memory lives

Desktop: on your computer, open to read and edit. Cloud: on your EU-hosted server, encrypted with a key Ferrox Labs holds

On n8n Cloud, or on your server if you self-host

Compliance

Your files and memory stay on your computer, and the code is open to read

n8n aligns its security program to SOC 2 with annual independent audits; the SOC 2 report goes to enterprise customers

How it signs in

You sign in to your AI provider yourself; Murage doesn’t handle your Claude login

Credentials live in n8n; Enterprise adds external secret stores

Reach

Chat apps

Message your Chief of Staff on Telegram, Slack, Discord or WhatsApp

Chat and messaging nodes inside workflows, plus n8n’s built-in chat

Voice

Call any bot in the app and just talk; bots leave voice notes. Ask your Chief for a voice note in Telegram, Slack, Discord or WhatsApp, and on WhatsApp send it one of your own

Not described on n8n’s homepage

Working together

Bots run your n8n workflows through n8n’s MCP server, and a workflow can start a Murage task by webhook

n8n’s instance-level MCP server exposes the workflows you switch on

For your staff

Each person runs Murage on their own computer; export a team and everyone imports it

Unlimited users on every n8n Cloud plan

An n8n alternative where a team does the work

Keep your workflows. Let your bots run them.

Your n8n workflows become tools your team can use.

At a glance.

Where your work lives.

Who does the work.

What Murage adds for an n8n user.

Keep your workflows. Add a team on top

Describe the outcome, not the steps

Bots that do the work itself

Asks first, by default

Where n8n is ahead.

Which should you choose?

Choose Murage if

Choose n8n if

Every detail.

Getting in

Where it runs

Brains

The team

Hands

Control

Your data

Reach

Same job, both ways.

Send a weekly numbers report

Triage the support inbox

Research three competitors and write a brief

Sync invoices to your books every night

Start a bot’s job from a workflow

Already using n8n: your first day with Murage

Questions about Murage and n8n.

Get started in four steps

Download Murage

Pick an engine

Meet Ember

Give it the first job

Read next.

Give your first job to Murage.

Search murage.ai

An n8n alternative where a team does the work

Keep your workflows. Let your bots run them.

Your n8n workflows become tools your team can use.

At a glance.

Where your work lives.

Who does the work.

What Murage adds for an n8n user.

Keep your workflows. Add a team on top

Describe the outcome, not the steps

Bots that do the work itself

Asks first, by default

Where n8n is ahead.

Which should you choose?

Choose Murage if

Choose n8n if

Every detail.

Getting in

Where it runs

Brains

The team

Hands

Control

Your data

Reach

Same job, both ways.

Send a weekly numbers report

Triage the support inbox

Research three competitors and write a brief

Sync invoices to your books every night

Start a bot’s job from a workflow

Already using n8n: your first day with Murage

Questions about Murage and n8n.

Get started in four steps

Download Murage

Pick an engine

Meet Ember

Give it the first job

Read next.

Give your first job to Murage.