The n8n MCP has been enhanced with new tools that dramatically improve the AI agent experience when building n8n workflows. The key improvement is the get_node_essentials tool which reduces response sizes by 95% while providing all the information needed for basic configuration.
Purpose: Get only the 10-20 most important properties for a node instead of 200+
When to use:
- Starting to configure a new node
- Need quick access to common properties
- Want working examples
- Building basic workflows
Example usage:
{
"name": "get_node_essentials",
"arguments": {
"nodeType": "nodes-base.httpRequest"
}
}Response structure:
{
"nodeType": "nodes-base.httpRequest",
"displayName": "HTTP Request",
"description": "Makes HTTP requests and returns the response data",
"requiredProperties": [
{
"name": "url",
"displayName": "URL",
"type": "string",
"description": "The URL to make the request to",
"placeholder": "https://api.example.com/endpoint"
}
],
"commonProperties": [
{
"name": "method",
"type": "options",
"options": [
{ "value": "GET", "label": "GET" },
{ "value": "POST", "label": "POST" }
],
"default": "GET"
}
// ... 4-5 more common properties
],
"examples": {
"minimal": {
"url": "https://api.example.com/data"
},
"common": {
"method": "POST",
"url": "https://api.example.com/users",
"sendBody": true,
"contentType": "json",
"jsonBody": "{ \"name\": \"John\" }"
}
},
"metadata": {
"totalProperties": 245,
"isAITool": false,
"isTrigger": false
}
}Benefits:
- 95% smaller response (5KB vs 100KB+)
- Only shows properties you actually need
- Includes working examples
- No duplicate or confusing properties
- Clear indication of what's required
Purpose: Find specific properties within a node without downloading everything
When to use:
- Looking for authentication options
- Finding specific configuration like headers or body
- Exploring what options are available
- Need to configure advanced features
Example usage:
{
"name": "search_node_properties",
"arguments": {
"nodeType": "nodes-base.httpRequest",
"query": "auth"
}
}Response structure:
{
"nodeType": "nodes-base.httpRequest",
"query": "auth",
"matches": [
{
"name": "authentication",
"displayName": "Authentication",
"type": "options",
"description": "Method of authentication to use",
"path": "authentication",
"options": [
{ "value": "none", "label": "None" },
{ "value": "basicAuth", "label": "Basic Auth" }
]
},
{
"name": "genericAuthType",
"path": "genericAuthType",
"showWhen": { "authentication": "genericCredentialType" }
}
],
"totalMatches": 5,
"searchedIn": "245 properties"
}-
Start with essentials:
get_node_essentials("nodes-base.httpRequest") -
Use the provided examples:
- Start with
minimalexample - Upgrade to
commonfor typical use cases - Modify based on your needs
- Start with
-
Search for specific features (if needed):
search_node_properties("nodes-base.httpRequest", "header")
-
Get documentation first:
get_node_documentation("nodes-base.httpRequest") -
Get essentials for the basics:
get_node_essentials("nodes-base.httpRequest") -
Search for advanced properties:
search_node_properties("nodes-base.httpRequest", "proxy") -
Only use get_node_info if absolutely necessary:
get_node_info("nodes-base.httpRequest") // Last resort - 100KB+ response
// Start with essentials
const essentials = get_node_essentials("nodes-base.httpRequest");
// Use the POST example
const config = essentials.examples.common;
// Modify for your needs
config.url = "https://api.myservice.com/endpoint";
config.jsonBody = JSON.stringify({ my: "data" });// Get webhook essentials
const essentials = get_node_essentials("nodes-base.webhook");
// Start with minimal
const config = essentials.examples.minimal;
config.path = "my-webhook-endpoint";// Get database essentials
const essentials = get_node_essentials("nodes-base.postgres");
// Check available operations
const operations = essentials.operations;
// Use appropriate example
const config = essentials.examples.common;-
Always start with get_node_essentials - It has everything needed for 90% of use cases
-
Use examples as templates - They're tested, working configurations
-
Search before diving deep - Use search_node_properties to find specific options
-
Check metadata - Know if you need credentials, if it's a trigger, etc.
-
Progressive disclosure - Start simple, add complexity only when needed
The essentials tool has optimized configurations for 20+ commonly used nodes:
- Core: httpRequest, webhook, code, set, if, merge, splitInBatches
- Databases: postgres, mysql, mongodb, redis
- Communication: slack, email, discord
- Files: ftp, ssh, googleSheets
- AI: openAi, agent
- Utilities: executeCommand, function
For other nodes, the tool automatically extracts the most important properties.
Based on testing with top 10 nodes:
- Average size reduction: 94.3%
- Response time improvement: 78%
- Properties shown: 10-20 (vs 200+)
- Usability improvement: Dramatic
If you're currently using get_node_info, here's how to migrate:
const node = get_node_info("nodes-base.httpRequest");
// Parse through 200+ properties
// Figure out what's required
// Deal with duplicates and conditionalsconst essentials = get_node_essentials("nodes-base.httpRequest");
// Use essentials.requiredProperties
// Use essentials.commonProperties
// Start with essentials.examples.commonQ: The tool says node not found
A: Use the full node type with prefix: nodes-base.httpRequest not just httpRequest
Q: I need a property that's not in essentials
A: Use search_node_properties to find it, or get_node_info as last resort
Q: The examples don't cover my use case A: Start with the closest example and modify. Use search to find additional properties.
Q: How do I know what properties are available?
A: Check metadata.totalProperties to see how many are available, then search for what you need
Planned enhancements:
- Task-based configurations (e.g., "post_json_with_auth")
- Configuration validation
- Property dependency resolution
- More node coverage
The new essentials tools make n8n workflow building with AI agents actually practical. Instead of overwhelming agents with hundreds of properties, we provide just what's needed, when it's needed. This results in faster, more accurate workflow creation with fewer errors.