feat(ai-builder): Prefer core nodes over code node

packages/@n8n/ai-workflow-builder.ee/src/prompts/agents/configurator.prompt.ts

@@ -55,6 +55,30 @@ Examples in parameter configuration:
 - "Set value to ={{ $('Previous Node').item.json.value }}"
 - "Set message to ={{ $('HTTP Request').item.json.message }}"`;
 
+const EXPRESSION_TECHNIQUES = `Expressions support JavaScript methods
+
+Regex operations:
+- Test pattern: ={{ /pattern/.test($json.text) }}
+- Extract match: ={{ $json.text.match(/pattern/)?.[1] }}
+- Replace text: ={{ $json.text.replace(/pattern/, 'replacement') }}
+- Split by pattern: ={{ $json.text.split(/pattern/) }}
+
+String operations:
+- Uppercase: ={{ $json.text.toUpperCase() }}
+- Trim whitespace: ={{ $json.text.trim() }}
+- Substring: ={{ $json.text.substring(0, 10) }}
+
+Array operations:
+- First item: ={{ $json.items[0] }}
+- Filter: ={{ $json.items.filter(i => i.active) }}
+- Map: ={{ $json.items.map(i => i.name) }}
+- Join: ={{ $json.items.join(', ') }}
+
+Generating items from expressions (use with Split Out node):
+- Create array from comma string: ={{ $json.text.split(',') }}
+- Generate range: ={{ Array.from({{length: 5}}, (_, i) => i + 1) }}
+- Use with Split Out node to create multiple output items from a single input`;
+
 const TOOL_NODE_EXPRESSIONS = `Tool nodes (types ending in "Tool") support $fromAI expressions:
 - "Set sendTo to ={{ $fromAI('to') }}"
 - "Set subject to ={{ $fromAI('subject') }}"
@@ -169,6 +193,7 @@ export function buildConfiguratorPrompt(): string {
 		.section('workflow_json_detection', WORKFLOW_JSON_DETECTION)
 		.section('parameter_configuration', PARAMETER_CONFIGURATION)
 		.section('data_referencing', DATA_REFERENCING)
+		.section('expression_techniques', EXPRESSION_TECHNIQUES)
 		.section('tool_node_expressions', TOOL_NODE_EXPRESSIONS)
 		.section('critical_parameters', CRITICAL_PARAMETERS)
 		.section('default_values_warning', DEFAULT_VALUES_WARNING)

packages/@n8n/ai-workflow-builder.ee/src/prompts/agents/discovery.prompt.ts

@@ -218,6 +218,65 @@ const SUB_NODES_SEARCHES = `When searching for AI nodes, ALSO search for their r
 
 const STRUCTURED_OUTPUT_PARSER = structuredOutputParser.usage;
 
+const CODE_NODE_ALTERNATIVES = `CRITICAL: Prefer native n8n nodes over Code node. Code nodes are slower (sandboxed environment).
+
+**Edit Fields (Set) with expressions is your go-to node for data manipulation:**
+- Adding new fields with values or expressions
+- Renaming or removing fields
+- Mapping data from one structure to another
+- Massaging/transforming field values using expressions
+- Restructuring JSON objects
+- Setting variables or constants
+- Creating hardcoded values, lists and objects
+
+**Native node alternatives - use INSTEAD of Code node:**
+
+| Task | Use This |
+|------|----------|
+| Add/modify/rename fields | Edit Fields (Set) |
+| Set hardcoded values/config/objects/lists | Edit Fields (Set) |
+| Map/massage/transform data | Edit Fields (Set) |
+| Generate list of items | Edit Fields (Set) + Split Out |
+| Filter items by condition | Filter |
+| Route by condition | If or Switch |
+| Split array into items | Split Out |
+| Combine multiple items | Aggregate |
+| Merge data from branches | Merge |
+| Summarize/pivot data | Summarize |
+| Sort items | Sort |
+| Remove duplicates | Remove Duplicates |
+| Limit items | Limit |
+| Format as HTML | HTML |
+| Parse AI output | Structured Output Parser |
+| Date/time operations | Date & Time |
+| Compare datasets | Compare Datasets |
+| Throw errors | Stop and Error |
+| Regex pattern matching | If node with expression |
+| Extract text with regex | Edit Fields (Set) with expression |
+| Validate text format | If node with regex expression |
+| Parse/extract fields from text | Edit Fields (Set) |
+
+**Regex works in expressions - no Code node needed:**
+- Test pattern
+- Extract match
+- Replace text
+- Split by pattern
+
+**Code node is ONLY appropriate for:**
+- Complex multi-step algorithms that cannot be expressed in single expressions
+- Operations requiring external libraries or complex data structures
+- Mathematical calculations beyond simple expressions
+
+**NEVER use Code node for:**
+- Simple data transformations (use Edit Fields)
+- Setting hardcoded values or configuration (use Edit Fields)
+- Filtering/routing (use Filter, If, Switch)
+- Array operations (use Split Out, Aggregate)
+- Basic data restructuring (use Edit Fields + expressions)
+- Regex operations (use expressions in If or Edit Fields nodes)
+- Text extraction or parsing (use Edit Fields with expressions)
+- Logging using console.log unless user explicitly asks - only useful for debugging, not production`;
+
 const CRITICAL_RULES = `- NEVER ask clarifying questions
 - ALWAYS call get_best_practices first
 - THEN Call search_nodes to learn about available nodes and their inputs and outputs
@@ -227,7 +286,8 @@ const CRITICAL_RULES = `- NEVER ask clarifying questions
 - ONLY flag connectionChangingParameters if they appear in <input> or <output> expressions
 - If no parameters appear in connection expressions, return empty array []
 - Output ONLY: nodesFound with {{ nodeName, version, reasoning, connectionChangingParameters }}
-- When user specifies a model name (e.g., 'gpt-4.1-mini') try to use this if it is a valid option`;
+- When user specifies a model name (e.g., 'gpt-4.1-mini') try to use this if it is a valid option
+- PREFER native n8n nodes (especially Edit Fields) over Code node`;
 
 const RESTRICTIONS = `- Output text commentary between tool calls
 - Include bestPractices or categorization in submit_discovery_results
@@ -289,6 +349,7 @@ export function buildDiscoveryPrompt(options: DiscoveryPromptOptions): string {
 		.section('process', processSteps)
 		.section('technique_categorization', TECHNIQUE_CATEGORIZATION)
 		.section('technique_clarifications', TECHNIQUE_CLARIFICATIONS)
+		.section('code_node_alternatives', CODE_NODE_ALTERNATIVES)
 		.section('connection_changing_parameters', CONNECTION_PARAMETERS)
 		.section('dynamic_output_nodes', DYNAMIC_OUTPUT_NODES)
 		.section('sub_nodes_searches', SUB_NODES_SEARCHES)

packages/@n8n/ai-workflow-builder.ee/src/tools/best-practices/data-transformation.ts

@@ -69,22 +69,6 @@ export class DataTransformationBestPractices implements BestPracticesDocument {
 
 #### Code Node (n8n-nodes-base.code)
 
-**Built-in Nodes vs. Code Node**
-- Prefer basic built-in nodes (Edit Fields, Filter, Split Out, Summarize, Aggregate, etc.) over Code node. Use Code only for complex logic that can't be achieved otherwise.
-- Rule of thumb: if the goal can be achieved with fewer than 5 basic nodes, use basic nodes
-
-**When NOT to Use**: Code node may be slower than core nodes (like Edit Fields, If, Switch, Split Out, Aggregate, etc.) as Code nodes run in a sandboxed environment. Avoid the code node where possible — it should only be used for complex transformations that can't be done with other nodes. For example, DO NOT use it for:
-- Adding or removing fields from items (use the 'edit fields' node instead)
-- Single-line data transformations of item fields (use the 'edit fields' node instead)
-- Filtering items based on their fields (use the 'filter' node instead)
-- Pivoting or summarizing data across multiple items (use the 'summarize' node instead)
-- Splitting arrays inside items out into multiple items (use the 'split out' node instead)
-- Aggregating multiple items into a single item (use the 'aggregate' node instead)
-- Sorting items in an array based on their fields (use the 'Sort' node instead)
-- Generating HTML from text or formatting text as HTML (use the 'HTML' node set to operation 'Generate HTML Template' or 'Convert to HTML Table' instead)
-
-**When to Use**: Complex transformations impossible with built-in nodes
-
 **Execution Modes**:
 - "Run Once per Item": Process each item independently
 - "Run Once for All Items": Access entire dataset (for aggregation)