feat: Decrease AI Builder wordiness (no-changelog) (#18920)

packages/@n8n/ai-workflow-builder.ee/evaluations/README.md

@@ -161,6 +161,7 @@ The AI Workflow Builder agent needs access to node definitions to generate workf
 1. Run your n8n instance
 2. Download the node definitions from locally running n8n instance(http://localhost:5678/types/nodes.json)
 3. Save the node definitions to `evaluations/nodes.json`
+` curl -o evaluations/nodes.json http://localhost:5678/types/nodes.json`
 
 The evaluation will fail with a clear error message if `nodes.json` is missing.
 

packages/@n8n/ai-workflow-builder.ee/src/tools/add-node.tool.ts

@@ -13,7 +13,7 @@ import { findNodeType } from './helpers/validation';
 import type { AddedNode } from '../types/nodes';
 import type { AddNodeOutput, ToolError } from '../types/tools';
 
-const DISPLAY_TITLE = 'Adding node';
+const DISPLAY_TITLE = 'Adding nodes';
 
 /**
  * Schema for node creation input
@@ -77,7 +77,8 @@ function getCustomNodeTitle(
 		}
 	}
 
-	return DISPLAY_TITLE;
+	// single "node" not plural "nodes" because this pertains to this specific tool call
+	return 'Adding node';
 }
 
 /**

packages/@n8n/ai-workflow-builder.ee/src/tools/prompts/main-agent.prompt.ts

@@ -3,26 +3,21 @@ import { ChatPromptTemplate } from '@langchain/core/prompts';
 import { instanceUrlPrompt } from '@/chains/prompts/instance-url';
 
 const systemPrompt = `You are an AI assistant specialized in creating and editing n8n workflows. Your goal is to help users build efficient, well-connected workflows by intelligently using the available tools.
-
-<prime_directive>
-ALWAYS end your workflow mutation responses with a brief note that the workflow can be adjusted if needed. For example: "Feel free to let me know if you'd like to adjust any part of this workflow!" This is mandatory for all workflow mutation responses.
-</prime_directive>
-
 <core_principle>
 After receiving tool results, reflect on their quality and determine optimal next steps. Use this reflection to plan your approach and ensure all nodes are properly configured and connected.
 </core_principle>
 
 <communication_style>
-Be warm, helpful, and most importantlyconcise. Focus on actionable information.
-- Lead with what was accomplished
-- Highlight only critical configuration needs
-- Provide clear next steps
-- Save detailed explanations for when users ask
-- One emoji per section maximum
+Keep responses concise.
+
+CRITICAL: Do NOT provide commentary between tool calls. Execute tools silently.
+- NO progress messages like "Perfect!", "Now let me...", "Excellent!"
+- NO descriptions of what was built or how it works
+- NO workflow features or capabilities explanations
+- Only respond AFTER all tools are complete
+- Response should only contain setup/usage information
 </communication_style>
 
-<tool_execution_strategy>
-For maximum efficiency, invoke all relevant tools simultaneously when performing independent operations. This significantly reduces wait time and improves user experience.
 
 Parallel execution guidelines:
 - ALL tools support parallel execution, including add_nodes
@@ -316,7 +311,7 @@ When unsure about specific values:
 - Add nodes and connections confidently
 - For uncertain parameters, use update_node_parameters with clear placeholders
 - For tool nodes with dynamic values, use $fromAI expressions instead of placeholders
-- Always mention what needs user input in your response
+- Always mention what needs user to configure in the setup response
 
 Example for regular nodes:
 update_node_parameters({{
@@ -329,41 +324,30 @@ update_node_parameters({{
   nodeId: "gmailTool1",
   instructions: ["Set sendTo to {{ $fromAI('to') }}", "Set subject to {{ $fromAI('subject') }}"]
 }})
-
-Then tell the user: "I've set up the Gmail Tool node with dynamic AI parameters - it will automatically determine recipients and subjects based on context."
 </handling_uncertainty>
 
 `;
 
 const responsePatterns = `
 <response_patterns>
-After completing workflow tasks, follow this structure:
+IMPORTANT: Only provide ONE response AFTER all tool execution is complete.
 
-1. **Brief Summary** (1-2 sentences)
-   State what was created/modified without listing every parameter
+Response format:
+**⚙️ How to Setup** (numbered format)
+- List credentials and parameters that need to configured
+- Only list incomplete tasks that need user action (skip what's already configured)
 
-2. **Key Requirements** (if any)
-   - Credentials needed
-   - Parameters the user should verify
-   - Any manual configuration required
+**ℹ️ How to Use**
+- Only essential user actions (what to click, where to go)
 
-3. **How to Use** (when relevant)
-   Quick steps to get started
+End with: "Let me know if you'd like to adjust anything."
 
-4. **Next Steps** (if applicable)
-   What the user might want to do next
-
-<communication_style>
-Be warm, helpful, and most importantly concise. Focus on actionable information.
-- Lead with what was accomplished
-- Provide clear next steps
-- Highlight only critical configuration needs
-- Be warm and encouraging without excessive enthusiasm
-- Use emojis sparingly (1-2 max per response)
-- Focus on what the user needs to know
-- Expand details only when asked
-- End with a brief note that the workflow can be adjusted if needed
-</communication_style>
+ABSOLUTELY FORBIDDEN IN BUILDING MODE:
+- Any text between tool calls
+- Progress updates during execution
+- "Perfect!", "Now let me...", "Excellent!"
+- Describing what was built
+- Explaining workflow functionality
 </response_patterns>
 `;
 

packages/@n8n/ai-workflow-builder.ee/src/utils/stream-processor.ts

@@ -59,6 +59,7 @@ export function processStreamChunk(streamMode: string, chunk: unknown): StreamOu
 		if ((agentChunk?.compact_messages?.messages ?? []).length > 0) {
 			const lastMessage =
 				agentChunk.compact_messages!.messages![agentChunk.compact_messages!.messages!.length - 1];
+
 			const messageChunk: AgentMessageChunk = {
 				role: 'assistant',
 				type: 'message',
@@ -83,13 +84,17 @@ export function processStreamChunk(streamMode: string, chunk: unknown): StreamOu
 					content = lastMessage.content;
 				}
 
-				const messageChunk: AgentMessageChunk = {
-					role: 'assistant',
-					type: 'message',
-					text: content,
-				};
+				if (content) {
+					const messageChunk: AgentMessageChunk = {
+						role: 'assistant',
+						type: 'message',
+						text: content,
+					};
 
-				return { messages: [messageChunk] };
+					return { messages: [messageChunk] };
+				}
+
+				return null;
 			}
 		}
 

packages/@n8n/ai-workflow-builder.ee/src/utils/test/stream-processor.test.ts

@@ -86,6 +86,18 @@ describe('stream-processor', () => {
 				expect(message.text).toBe('Last message to display');
 			});
 
+			it('should handle compact_messages with empty content', () => {
+				const chunk = {
+					agent: {
+						messages: [{ content: 'First message' }, { content: [{ type: 'text', text: '' }] }],
+					},
+				};
+
+				const result = processStreamChunk('updates', chunk);
+
+				expect(result).toEqual(null);
+			});
+
 			it('should handle process_operations with workflow update', () => {
 				const workflowData = {
 					nodes: [{ id: 'node1', name: 'Test Node' }],