n8n Assistant Launches as a Preview AI Workflow Builder for Cloud and Docker Users

n8n has launched n8n Assistant, a preview AI-assisted workflow builder designed to turn a plain-language automation request into a workflow that teams can inspect, edit and debug in the n8n canvas. The release is available for new n8n Cloud instances by default and for eligible self-hosted Docker deployments running n8n 2.36 or later.

The central proposition is not simply generating a workflow template. n8n Assistant is intended to plan a workflow, build it on the canvas, request credentials when they are needed, execute the result, inspect the execution and propose fixes if something fails. The resulting automation remains a standard n8n workflow rather than a separate black-box artifact. That means teams retain access to the canvas, execution history and node-level logs.

According to n8n's official Assistant announcement, the feature is being shipped behind a preview flag and remains in active development. That status matters: n8n explicitly advises users to review workflows and execution logs before relying on an initial build. The assistant can iterate on a failed workflow, but it cannot eliminate the need to verify credentials, third-party configuration and the actual business logic being automated.

What n8n Assistant changes in workflow building

Traditional visual automation builders require a user to translate a process into triggers, nodes, conditions, credentials and mappings manually. n8n Assistant adds a conversational starting point. A user can describe an intended process, then work from the workflow the assistant creates in the existing editor.

That can reduce the blank-page problem for teams that know the outcome they need but are less familiar with the sequence of nodes required to build it. For example, a team may begin with a request to route incoming information through a series of connected services, then inspect and refine the generated workflow before using it operationally. The key boundary is that the assistant helps construct and troubleshoot the workflow, while the team still owns the decisions behind it.

The release emphasizes several connected capabilities:

• Plain-language workflow requests that the assistant translates into a planned automation.
• Canvas-based workflow creation, so the generated workflow can be edited directly in n8n.
• Credential requests during setup where a workflow needs access to connected services.
• Execution observation and per-node logs showing inputs, outputs and errors.
• Iterative troubleshooting, where the assistant can propose a fix and rerun the workflow until it completes.

This transparency is a meaningful distinction from a purely conversational automation experience. A successful run is useful, but the ability to see what each node received, returned or failed on is what lets a team assess whether a workflow is actually doing the right thing.

Debugging remains part of the workflow

n8n Assistant can execute the workflow it has built and use execution information to help identify failures. However, an automation can fail for reasons outside the generated workflow structure. Missing credentials, incorrect permissions, service-specific setup and inaccurate assumptions about a process can all require human intervention.

For businesses, that makes the assistant most useful as a faster route to a visible first version of an automation, not as a reason to skip testing. Teams should review the workflow on the canvas, validate node behavior in the logs and confirm that the automation handles real operating conditions before making it part of a critical process.

Cloud and self-hosted availability

n8n's initial availability differs by deployment model. New Cloud instances receive the assistant by default, except enterprise Cloud instances, which are not included in that default at present. Self-hosted users need to run version 2.36 or later and provide their own keys and environment variables. Self-hosted npm installations are not supported at this time, while enterprise support is on the roadmap.

The setup requirements are important for self-hosted teams. The assistant is not a feature that automatically appears in every existing installation. Organizations running Docker need the required version and configuration, while those using npm will need to wait for expanded support or use a supported deployment path.

Cost, practical value and preview limits

n8n describes Assistant usage as AI credit-based. Costs are tied to the number of debugging rounds and scale with the user's plan. The announcement does not provide a fixed universal price in the supplied information, so teams evaluating the feature should treat repeated iteration as a factor in usage costs rather than assuming every generated workflow will cost the same.

The practical value is clearest where workflow design time is a bottleneck. A manager or operations team can move from an automation objective to an editable draft more quickly, then use the platform's execution records to validate it. That can make experimentation more approachable, particularly for repeatable tasks that connect existing business tools. It does not remove the work of defining a sound process, supplying access credentials or checking the output.

Because this is a preview release, the right operating approach is deliberate. Start with contained workflows, examine node-level inputs and outputs, and keep a human review step for early builds. The assistant's ability to rerun and refine a workflow is helpful, but n8n's own guidance is clear that the first attempt may not always be correct.

For companies looking to use AI to reduce manual automation design without losing technical visibility, Scalevise can help turn promising workflows into reliable operational systems. Our n8n setup service can help with deployment, integrations, workflow design and the practical checks needed before automations support day-to-day work. This is especially useful when credentials, connected services and error handling need to work together rather than exist as a prototype. Discuss your n8n automation project with Scalevise.

Frequently Asked Questions

What is n8n Assistant?

n8n Assistant is a preview AI-assisted workflow builder in n8n. It can plan and build a workflow from a plain-language request, request credentials, run the workflow, observe execution and help iterate on failures.

Are n8n Assistant workflows editable?

Yes. Workflows created by the assistant are standard n8n workflows that teams can edit on the canvas. They also retain execution history and per-node logs.

Who can use n8n Assistant at launch?

It is enabled by default for new n8n Cloud instances, except enterprise Cloud instances for now. Self-hosted Docker deployments can use it with n8n 2.36 or later plus user-provided keys and environment variables. Self-hosted npm installs are not supported currently.

How does n8n Assistant pricing work?

n8n describes the feature as AI credit-based. Costs depend on the number of debugging rounds and scale with the user's plan.

Should teams trust an Assistant-generated workflow without review?

No. n8n says initial builds may not always be correct. Teams should check the workflow, execution history and per-node logs before relying on it.

Conclusion

n8n Assistant adds an AI-guided path from an automation idea to a visible, executable n8n workflow. Its strongest early advantage is the combination of assisted building with editable workflows and transparent logs. With preview status, deployment limits and usage-based credits, teams should begin with reviewable use cases and treat testing as part of implementation.