n8n Changelog 2026: Key Updates for Workflow Automation Teams

Staying current with n8n's release cycle is essential for teams building robust workflow automation. This changelog highlights the most impactful updates shipped in 2026, focusing on features that expand capabilities, enhance control, and improve observability. Each entry is tied to its release version, listed newest first, and written to stand alone for easy sharing.

For a complete view, pair this page with n8n's other release resources:

• Release notes: Comprehensive listing of all feature-level updates per release.
• GitHub releases: Detailed change logs linked to commits, including bug fixes and minor improvements.
• Archived notes: Historical pages for 2.x, 1.x, and 0.x remain available for reference.

Stable vs. Beta: Choosing the Right Version

n8n ships a new minor version most weeks. The stable version is production-ready, while beta offers the latest features but may be unstable. As of this writing, the current stable is 2.35.6, and the current beta is 2.36.4. For production workflow automation, always pin to stable. Use beta in sandbox environments to test upcoming capabilities and provide feedback via the n8n community forum.

Return Webhook Responses of Any Size from Workers

Released: 2026-08-04 in n8n 2.34

The Challenge with Large Webhook Responses

In queue mode, n8n workers execute workflows, but the client that initiated the webhook request remains connected to the main or webhook instance. When a Respond to Webhook node sends a response, it must travel back through the queue. Redis holds the entire response while in transit, which is why n8n historically capped it at 64 MiB by default (configurable via N8N_WEBHOOK_RESPONSE_RELAY_SIZE_MAX).

Any response exceeding this limit would fail the node. Common scenarios include:

• A database query returning a wide result set
• Aggregated responses from multiple API calls
• Dynamically assembled CSV or XML documents

The Solution: Binary Data Offloading

With n8n 2.34, you can enable response offloading by setting N8N_WEBHOOK_RESPONSE_RELAY_OFFLOAD_ENABLED=true on your workers. When enabled, response bodies exceeding the limit are stored in binary data storage instead of Redis. The queue message carries only a reference, and the main instance streams the body from storage to the client. Once delivered, n8n deletes the stored copy, ensuring no data accumulation.

Key benefits:

• Redis memory stays flat regardless of response size
• Streaming support for large payloads without buffering
• Automatic cleanup prevents storage bloat

Configuration Requirements

Before enabling offloading, ensure:

1. Binary data mode is set to a storage-backed mode (anything except default). Recommended: s3 or azure, as both support chunked streaming.
2. Shared storage accessible by all instances (main, webhook, and workers).
3. Main instance runs n8n 2.34.0 or later to read offloaded bodies. The variable ships disabled to prevent compatibility issues—upgrade your main and webhook instances first, then enable on workers.

Expert Tip: When to Use Offloading

For most workflows, the 64 MiB default is sufficient. Enable offloading only when you anticipate responses exceeding this threshold, such as exporting large datasets or generating bulky reports. Monitor Redis memory usage to identify if offloading is necessary.

For full configuration details and troubleshooting, refer to the Large webhook responses documentation.

Read and Write SharePoint Excel Workbooks Directly

Released: 2026-07-28 in n8n 2.33

Direct Access to SharePoint Data

The new Microsoft Excel (SharePoint) node enables reading and writing Excel workbooks stored in SharePoint document libraries, including files shared via Microsoft Teams sites. This eliminates the need for separate authentication or intermediate steps to fetch file contents, streamlining workflow automation for organizations heavily using SharePoint.

Authentication Methods

The node supports two authentication approaches:

1. User-based (OAuth2): Sign in as a person using a Microsoft OAuth2 credential with Sites.ReadWrite.All (or Sites.Read.All for read-only).
2. App-based (Service Principal): Use a Microsoft Entra Service Principal credential for unattended workflows that require no user interaction—ideal for background automation.

Practical Use Cases

• Automated reporting: Pull data from SharePoint-hosted Excel files to generate dashboards.
• Data synchronization: Sync changes between SharePoint workbooks and external databases.
• Team collaboration: Trigger workflows when a workbook is updated, then process the new data.

Expert Insight: Choosing Authentication

For workflows that run on a schedule or in response to events (e.g., webhooks), app-based authentication is more reliable because it doesn't depend on a user's session. Use user-based OAuth2 when you need to respect user-specific permissions or when the workflow is initiated by a user action.

Learn more in the Microsoft Excel (SharePoint) node documentation.

Capture Who Approved and When in Human-in-the-Loop Steps

Released: 2026-07-07 in n8n 2.30

Full Audit Trail for Approvals

Every Send and Wait node across Slack, Telegram, Discord, WhatsApp, Google Chat, Gmail, Outlook, Email/SMTP, Microsoft Teams, and the Chat Trigger node now includes a respondedAt ISO-8601 timestamp in its output when a response is received. This field appears automatically alongside existing fields like approved, text, or form data, with no configuration required and no change to the output shape for existing workflows.

Advanced Interactivity for Slack and Telegram

For Slack and Telegram, the new Advanced Interactivity options on the Send and Wait for Response operation take approvals further:

• One-tap approvals: Approvers respond directly within the app.
• Respondent details: The node output records who responded—ID, name, username, and (for Slack, when scopes allow) email—plus channel and message ID.
• Restrict approvers: Limit who can approve by listing user IDs in Restrict Who Can Approve. Unauthorized users receive a private notice you customize, and the workflow continues waiting.
• Post-decision behavior: Control what happens to the message via the After Decision setting: show the outcome and remove buttons (default), remove buttons only, or leave the message unchanged.

Setup Requirements

For Slack:

• Your n8n instance must be publicly reachable over HTTPS.
• Enable Interactivity in your Slack app.
• Set the Request URL to https://<your-n8n-instance>/webhook-waiting-slack.
• Paste your app's signing secret into the Signature Secret field of your Slack credential.
• In the Slack node, set Response Type to Approval and enable Capture Who Responded.

For Telegram:

• Your instance must be reachable over HTTPS on ports 443, 80, 88, or 8443.
• Enable Approve Within Chat in the Advanced Interactivity section. n8n registers the webhook automatically using your existing Telegram credential—no extra setup on Telegram's side.

Expert Tip: Designing Approval Workflows

When building human-in-the-loop automation, always define a fallback for unapproved requests. For example, if no one approves within a set timeframe, automatically escalate or notify a manager. Use the respondedAt timestamp to track response times and identify bottlenecks in your approval processes.

For detailed guidance, see the Approvals in Slack documentation.

App-Only Authentication for Microsoft Nodes

Released: 2026-07-07 in n8n 2.30

Breaking Free from User Sessions

Microsoft nodes now support app-only authentication using a Microsoft Entra Service Principal. This allows workflows to run as an application rather than a signed-in user, eliminating dependencies on individual user accounts.

OneDrive and Outlook gained this option in n8n 2.29, followed by Excel 365, Microsoft Teams, and Microsoft To Do in 2.30—all sharing a single app-only credential.

Why This Matters

Previously, Microsoft automations were tied to a person's OAuth session. When that person left the company or their token expired, the workflow broke. With app-only authentication:

• Non-interactive authentication with tenant-level permissions
• Target specific resources: user mailboxes, drives, or sites
• Unattended operation: read a shared mailbox, process files in any user's drive, or post to Teams channels without anyone logged in

Migration and Compatibility

• OAuth2 remains the default for all nodes, so existing workflows are unaffected.
• Operations that require a signed-in user are disabled per node with a clear error message.
• OneDrive and Outlook support shipped in n8n 2.29 (2026-06-30).

Expert Insight: Security Considerations

When using app-only authentication, grant the minimum permissions required for your workflows. Use the principle of least privilege to limit the service principal's access. Regularly review and rotate client secrets to maintain security.

mTLS Authentication for Kafka

Released: 2026-07-07 in n8n 2.30

The Kafka credential now supports mutual TLS (mTLS), allowing you to provide a CA certificate, client certificate, and private key (PEM) to connect to brokers that require client-certificate authentication. This applies to the Kafka node, the Kafka Trigger, and the credential test. n8n validates that the certificate and key match before saving.

Why mTLS Matters for Workflow Automation

mTLS provides a higher level of security than standard TLS by authenticating both the client and server. This is critical for enterprise environments where Kafka brokers are configured to accept connections only from verified clients.

Implementation Tips

• Ensure your certificates are in PEM format.
• Store certificates securely, ideally in n8n's credential vault.
• Test the connection using the credential test feature before deploying workflows.

AI Assistant: Describe a Goal, Get a Working Automation

Released: 2026-07-09 in n8n 2.29.9

The New Way to Build Workflows

The AI Assistant lets you describe an automation in plain language, and it will plan, build, test, and iterate until the workflow runs successfully. Access it from anywhere in your instance or expand it into a side-by-side view with the workflow canvas.

How It Works

1. Describe your goal: Tell the assistant what you want to automate, or start from a suggested prompt.
2. Receive a structured plan: The assistant proposes a step-by-step plan and asks clarifying questions.
3. Build and execute: It constructs the workflow in your selected project, executes it as it goes, and fixes errors it encounters.

Practical Applications

• Rapid prototyping: Quickly generate a working workflow for a new use case.
• Learning aid: See how n8n structures complex automations.
• Productivity boost: Reduce time spent on repetitive workflow setup.

Expert Tip: Guiding the AI Assistant

Be as specific as possible when describing your automation. Include details like data sources, trigger conditions, and desired outputs. The more context you provide, the more accurate the generated workflow will be. Always review and test the AI-generated workflow before deploying to production.

AI Assistant supersedes the older AI Workflow generation features, offering a more interactive and iterative approach.

Best Practices for Leveraging n8n Updates

To maximize the value of these new capabilities in your workflow automation:

1. Plan upgrades strategically: Test new features in a staging environment before rolling out to production.
2. Monitor release notes: Subscribe to n8n's release notifications to stay informed about upcoming changes.
3. Leverage app-only auth: For long-running or scheduled workflows, migrate to app-only authentication to avoid breakage.
4. Implement approval workflows: Use the new audit trail features to enhance governance and compliance.
5. Experiment with AI Assistant: Use it to accelerate workflow development, but always validate outputs.

Conclusion

n8n's 2026 releases bring significant enhancements for workflow automation teams: unlimited webhook responses, direct SharePoint integration, robust approval tracking, app-only Microsoft authentication, mTLS for Kafka, and an AI-powered assistant. By understanding and adopting these features, you can build more resilient, scalable, and intelligent automations.

For the latest updates, always refer to the official n8n changelog and release notes.