UI display for N8N AI Agent tool usage in n8n.py (N8P Pipeline)

[owndev]

This pull request introduces a major update to the N8N pipeline integration with Open WebUI, focusing on a new AI Agent Tool Usage Display feature (v2.2.0), improved documentation, and enhancements to the example N8N workflows. The changes provide better transparency into agent actions, configurable display options for tool calls, and expanded support for different language models and tools.

Key changes:

1. AI Agent Tool Usage Display Feature

• Introduced a new feature that extracts and displays tool calls from N8N AI Agent workflows in non-streaming mode, with three verbosity levels (minimal, compact, detailed) and customizable length limits. This adds transparency and debugging capabilities for agent actions. [1] [2] [3] [4]
• Added a dedicated documentation file docs/n8n-tool-usage-display.md detailing configuration, limitations, and usage of the new tool usage display feature.

2. Documentation Improvements

• Updated README.md and docs/n8n-integration.md to describe the new tool usage display, its configuration, and its current limitation to non-streaming mode due to N8N's streaming implementation. [1] [2] [3]
• Added links to the new documentation and clarified integration steps and feature usage. [1] [2]

3. N8N Example Workflow Enhancements

• Updated Open_WebUI_Test_Agent.json and Open_WebUI_Test_Agent_Streaming.json workflows to:
  • Add support for Google Gemini and Think nodes, expanding the range of supported models and tools. [1] [2]
  • Change webhook response to JSON and ensure intermediateSteps are included in responses, enabling tool usage display. [1] [2]
  • Update model names and positions for better workflow clarity and testing. [1] [2] [3] [4] [5] [6] [7] [8] [9]
  • Add configuration for memory buffer and streaming options in the streaming workflow. [1] [2]

4. Workflow Metadata and Connectivity

• Updated workflow metadata, version IDs, and connection mappings to reflect the new nodes and ensure correct tool/model linkage in the workflow templates. [1] [2]

These changes significantly improve the observability, flexibility, and extensibility of the N8N pipeline integration with Open WebUI.

---

Closes: #75

[Copilot]

Pull Request Overview

This pull request introduces major enhancements to the N8N pipeline integration with Open WebUI (v2.2.0), focusing on AI Agent tool usage display with configurable verbosity levels, improved documentation, and updates to example workflows.

• Enhanced tool usage display feature with three verbosity levels (minimal, compact, detailed) and customizable length limits for non-streaming mode
• Comprehensive documentation updates including new dedicated tool usage guide
• Updated example N8N workflows with Google Gemini support and improved response format

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
pipelines/n8n/n8n.py	Added tool usage display functionality, new configuration valves, and enhanced response parsing logic
pipelines/n8n/Open_WebUI_Test_Agent_Streaming.json	Updated streaming workflow with Google Gemini support and enhanced configuration
pipelines/n8n/Open_WebUI_Test_Agent.json	Updated non-streaming workflow with JSON responses and intermediateSteps support
docs/n8n-tool-usage-display.md	New comprehensive documentation for tool usage display feature
docs/n8n-integration.md	Updated integration documentation with tool display information
README.md	Updated with tool usage display feature highlights and documentation links

Comments suppressed due to low confidence (1)

pipelines/n8n/n8n.py:1

• The model name 'gpt-5-mini' appears to be invalid. OpenAI's current models follow the pattern 'gpt-4o-mini', 'gpt-4', etc. GPT-5 has not been released as of my knowledge cutoff.

"""

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}