n8n Changelog 2026: Key Updates for Building Better Workflow Automation

For technical teams building on n8n, every release is an opportunity to expand what's possible with workflow automation. This changelog highlights the most impactful updates shipped in 2026, focusing on new capabilities, enhanced control, and improved observability. Each entry is tied to its release version and designed to stand alone, making it easy to share the specific update your team has been waiting for.

Resource Guide: This page serves as a curated, narrative summary of major features. For a complete listing of feature-level updates, refer to the official Release Notes. For granular details including bug fixes and commit-level changes, see GitHub Releases. Archived notes for versions 2.x, 1.x, and 0.x remain available for historical reference.

Understanding n8n's Release Cadence: Stable vs. Beta

n8n ships a new minor version most weeks. Understanding the distinction between release channels is critical for production workflow automation:

• Stable Version: Recommended for production use. This is the most battle-tested release, having undergone additional scrutiny.
• Beta Version: The most recent release, containing the newest features. While feature-complete, it may contain unstable elements or undiscovered edge cases.

Current Versions:

• Stable: 2.35.6
• Beta: 2.36.4

Expert Tip: For mission-critical workflow automation, always test new features in a staging environment using the beta version before upgrading your production instance. This allows you to validate new nodes and capabilities against your specific use cases without risking downtime.

If you encounter issues, the n8n Community Forum is the official channel for reporting bugs and seeking assistance.

Return Webhook Responses of Any Size from Workers

Released: 2026-08-04 in n8n 2.34

The Challenge: The 64 MiB Webhook Response Ceiling

In n8n's queue mode, the architecture separates the main/webhook instance from worker nodes that execute the workflow. When a client makes a webhook request, the main instance accepts the connection, but the worker performs the execution. The response from a Respond to Webhook node must travel back through the message queue (typically Redis) to reach the waiting client.

This architecture introduced a hard limitation: Redis must hold the entire response payload while in transit. By default, n8n caps this at 64 MiB (configurable via N8N_WEBHOOK_RESPONSE_RELAY_SIZE_MAX).

The Problem: Failed Workflows on Large Payloads

Any response exceeding this limit would cause the node to fail. This was a significant bottleneck for several common workflow automation scenarios:

• Database Queries: A node returning a wide result set from a complex SQL query or a large MongoDB aggregation.
• API Aggregation: A workflow that batches multiple API calls and combines the results into a single response.
• File Generation: A workflow that assembles a large CSV, XML, or JSON document for download.

The Solution: Binary Data Offloading

n8n 2.34 introduces a mechanism to bypass this limit entirely. By setting N8N_WEBHOOK_RESPONSE_RELAY_OFFLOAD_ENABLED=true on your worker nodes, n8n automatically stores any response body exceeding the 64 MiB threshold in your configured binary data storage.

Here's how it works:

1. The worker serializes the response and detects it exceeds the limit.
2. Instead of placing the full payload in the queue message, n8n writes the body to binary data storage.
3. The queue message now carries only a lightweight reference to the stored data.
4. The main instance receives the reference, streams the body from storage directly to the waiting client.
5. n8n automatically deletes the stored copy after successful delivery, preventing storage accumulation.

Result: Redis memory usage remains flat regardless of response size, enabling truly scalable webhook responses.

Configuration Requirements & Best Practices

This feature requires specific configuration to work correctly:

• Binary Data Mode: You must set N8N_DEFAULT_BINARY_DATA_MODE to a mode that stores data (any mode except default). n8n recommends s3 or azure, as both support efficient chunked streaming of the body.
• Shared Storage: All instances (main, webhook, and workers) must have access to the same storage backend.
• Upgrade Order: This is critical. Only a main instance running n8n 2.34.0 or later can read an offloaded body. The variable ships turned off by design. You must upgrade your main and webhook instances first, then enable the flag on your workers.

Example Use Case: A customer-facing reporting portal uses an n8n webhook to generate a custom CSV export of their transaction history. With offloading enabled, the workflow can now handle exports for enterprise clients with millions of rows, returning files that are hundreds of MiB in size without errors.

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

Read and Write SharePoint Excel Workbooks Directly in n8n

Released: 2026-07-28 in n8n 2.33

Direct Access to SharePoint Data

The new Microsoft Excel (SharePoint) node eliminates the friction of working with Excel files stored in SharePoint document libraries, including those shared via Microsoft Teams sites. Previously, accessing this data required a complex chain of nodes to authenticate, retrieve file content, and parse it. This node provides direct, native access to workbook data.

Two Authentication Methods for Flexibility

The node supports two distinct authentication methods to suit different operational needs:

1. User Authentication (OAuth2): Sign in as a person using a Microsoft OAuth2 credential. This requires the Sites.ReadWrite.All (or Sites.Read.All for read-only) scope. This is ideal for interactive workflows or when you want to leverage a user's existing permissions.

2. App-Only Authentication (Service Principal): Sign in as an application using a Microsoft Entra Service Principal credential. This is perfect for unattended, backend workflow automation that requires no user interaction and maintains access regardless of employee turnover.

Example Use Case: A finance team automates their monthly close process. A workflow triggered by a schedule uses the SharePoint node to read a budget workbook, perform calculations, and write updated figures back to the same file – all without any manual download/upload steps or user intervention.

For detailed setup instructions, see 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

Automatic Audit Trails 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 automatically includes a respondedAt ISO-8601 timestamp in its output. This provides a complete audit trail for every human-in-the-loop step in your workflow automation.

Key Features:

• Zero Configuration: The field appears automatically alongside existing approved, text, or form fields.
• Backward Compatible: The output shape for existing workflows remains unchanged, ensuring no breakage on upgrade.

Advanced Interactivity: Approvals Within Slack and Telegram

For Slack and Telegram, n8n 2.30 introduces Advanced Interactivity options that transform the approval experience:

• One-Tap Approvals: Approvers can respond directly within the app with a single tap or click.
• Rich Response Data: The node output records exactly who responded, including their ID, name, username, and (for Slack, when scopes allow) email, plus the channel and message ID.
• Restrict Who Can Approve: You can specify a list of user IDs allowed to approve. Anyone not on the list receives a private notice (with customizable wording), and the workflow continues to wait.
• Post-Decision Message Control: The After Decision setting lets you control the message state: show the outcome and remove buttons (default), remove buttons only, or leave the message unchanged.

Configuration for Slack Approvals

To enable this feature in Slack, you must:

1. Ensure your n8n instance is reachable from Slack over public HTTPS.
2. Enable Interactivity in your Slack app configuration.
3. Set the Request URL to https://<your-n8n-instance>/webhook-waiting-slack.
4. Paste your app's Signing Secret into the Signature Secret field of your Slack credential.
5. In the Slack node, set Response Type to Approval and enable Capture Who Responded.

Configuration for Telegram Approvals

For Telegram:

1. Your instance must be reachable over public HTTPS on a supported port (443, 80, 88, or 8443).
2. Enable Approve Within Chat in the same section. n8n handles webhook registration automatically using your existing Telegram credential – no additional setup on Telegram's side is required.

Example Use Case: A DevOps team uses a Slack-based approval workflow for production deployments. The new audit trail provides a clear record of who approved each release and when, satisfying compliance requirements without manual log-keeping.

Learn more in the Approvals in Slack documentation.

App-Only Authentication for Microsoft Nodes

Released: 2026-07-07 in n8n 2.30

Decoupling Workflows from User Sessions

A major pain point for Microsoft-centric workflow automation has been the reliance on individual user OAuth sessions. When an employee leaves the company or their token expires, the automation breaks. n8n 2.30 addresses this with app-only authentication using a Microsoft Entra Service Principal.

This allows workflows to run as an application with tenant-level permissions, targeting the specific user, mailbox, drive, or site you specify. This enables powerful, unattended automation scenarios:

• Read a Shared Mailbox: Process incoming emails from a department-level mailbox without needing a specific user's session.
• Process Files in Any User's Drive: Automate file operations across the organization, not just a single user's OneDrive.
• Post to Teams Channels: Send notifications and reports with nobody logged in.

Timeline and Compatibility

• n8n 2.29 (2026-06-30): OneDrive and Outlook gained app-only support.
• n8n 2.30: Excel 365, Microsoft Teams, and Microsoft To Do follow, all sharing a single app-only credential.

Important: OAuth2 remains the default authentication method, so existing workflows are untouched. Operations that only make sense for a signed-in user (e.g., accessing personal OneDrive folders) are disabled per node with a clear error message, preventing misconfiguration.

Expert Insight: This is a game-changer for IT teams managing large fleets of integrations. It centralizes security control under a single service principal, making it easier to audit access and revoke permissions globally if needed.

mTLS Authentication for Kafka

Released: 2026-07-07 in n8n 2.30

Enhanced Security for Event-Driven Architecture

The Kafka credential now supports mutual TLS (mTLS), a critical security feature for enterprise environments. You can now provide a CA certificate, client certificate, and private key (in PEM format) to connect to brokers that require client-certificate authentication.

Key Details:

• Comprehensive Coverage: mTLS applies to the Kafka node, the Kafka Trigger, and the credential test functionality.
• Validation: n8n validates that the certificate and key match before saving, preventing configuration errors.

Example Use Case: A financial services company processes real-time transaction data through Kafka. With mTLS support, n8n can now securely connect to their broker cluster, which requires client certificates for all connections, enabling a unified workflow automation platform for their event-driven architecture.

AI Assistant: Describe a Goal, Get a Working Automation

Released: 2026-07-09 in n8n 2.29.9

The Shift from Building to Describing

n8n's AI Assistant represents a paradigm shift in how workflow automation is created. Instead of manually dragging and configuring nodes, you can now describe an automation goal in plain language, and the AI Assistant will plan, build, test, and iterate on the workflow until it actually runs.

How It Works

1. Describe Your Goal: Open the chat from anywhere in your instance, or expand it into a side-by-side view with the workflow canvas.
2. AI Planning: The assistant proposes a structured plan and asks clarifying questions to understand your requirements.
3. Automated Building: It builds the workflow in your selected project, executing it as it goes.
4. Self-Correction: The AI identifies errors during execution and fixes them automatically.

You can start from scratch with your own description or use a suggested prompt. This feature supersedes the older AI Workflow generation tool, offering a more integrated and iterative experience.

Example Use Case: A marketing manager with no coding experience wants to automate lead scoring. They open the AI Assistant and type: "When a new lead is added to my CRM, score them based on their company size and job title, and send a Slack message to the sales team if the score is above 80." The AI Assistant handles the rest, creating a functional workflow in minutes.

Conclusion: Building More Resilient and Intelligent Workflow Automation

The 2026 releases from n8n demonstrate a clear focus on three pillars:

1. Scale and Reliability: The webhook response offloading feature removes a critical technical bottleneck, enabling more ambitious data-heavy workflows.
2. Enterprise-Grade Security and Governance: App-only authentication for Microsoft nodes and mTLS for Kafka provide the security controls required for enterprise adoption.
3. Intelligent Automation: The AI Assistant lowers the barrier to entry, making powerful workflow automation accessible to a broader audience while the enhanced audit trails provide the visibility technical teams need.

By leveraging these updates, teams can build more resilient, secure, and intelligent workflow automation that drives significant business value. Stay tuned for future releases as n8n continues to push the platform forward.

Frequently Asked Questions

What is the best way to get started with n8n Changelog 2026: New Features for Wor?

The best approach is to start with a clear goal in mind. Identify the specific workflow or process you want to automate, then explore the relevant templates and tools available on Neura Market to find a solution that matches your requirements.

How much does workflow automation typically cost?

Costs vary significantly depending on the platform and scale. Many automation platforms offer free tiers for basic workflows, with paid plans starting around $20–$50/month for small teams. Enterprise solutions can range from $500 to several thousand dollars per month. Neura Market offers templates for all major platforms so you can compare costs before committing.

Do I need technical skills to implement workflow automation?

Modern no-code and low-code platforms like Zapier, Make.com, and others have made automation accessible to non-technical users. Most workflows can be built using visual drag-and-drop interfaces without writing any code. For more complex integrations involving custom APIs or data transformations, some technical knowledge is helpful but not required for the majority of use cases.