n8n

​

Overview

​

Prerequisites

• ✅ n8n installed (self-hosted or cloud)
• ✅ Pylar project with published MCP tools
• ✅ MCP HTTP Stream URL and Bearer Token from Pylar

​

Step 1: Get Your Pylar Credentials

1. In Pylar, navigate to your project
2. Click “Publish” in the right sidebar
3. Copy both:
   • MCP HTTP Stream URL: https://mcp.publish.pylar.ai/mcp
   • Authorization Bearer Token: Your unique token

​

Step 2: Create HTTP Request Node

1. In n8n, create a new workflow
2. Add an “HTTP Request” node to your canvas
3. Connect it to your workflow trigger or previous node

​

Step 3: Configure HTTP Request Node

​

Basic Settings

1. Click on the HTTP Request node to configure it
2. Set the following:

https://mcp.publish.pylar.ai/mcp

​

Authentication

1. Expand the Authentication section
2. Select “Header Auth” or “Generic Credential Type”
3. Configure:
   • Name: Authorization
   • Value: Bearer YOUR_BEARER_TOKEN_HERE

​

Headers

• Content-Type: application/json

​

Request Body

1. Set Body Content Type to JSON
2. Add the JSON body for calling your MCP tool:

{
  "jsonrpc": "2.0",
  "id": "1",
  "method": "tools/call",
  "params": {
    "name": "fetch_engagement_scores_by_event_type",
    "arguments": {
      "event_type": "{{ $json.event_type }}"
    }
  }
}

​

Step 4: Select Tools

1. Select which tools you want the agent/workflow to have access to
2. Configure different nodes for different tools
3. Chain multiple tool calls if needed

​

Step 5: Test the Connection

1. Click “Execute Step” to test the node
2. Review the response to verify the tool works
3. Check that data is returned correctly

​

Step 6: Use in Your Workflow

1. Go to the chat in n8n and ask questions to the AI
2. Run the workflow to execute automation
3. The AI or workflow will use your Pylar tools to access data

​

Example Workflow

• Method: POST
• URL: https://mcp.publish.pylar.ai/mcp
• Auth: Bearer Token
• Body: Tool call with event_type from trigger

• Parse JSON response
• Extract engagement scores

• Check if score is above threshold

• Alert if score is high

​

Benefits

• ✅ Visual Workflows: Build automations visually
• ✅ No Code Required: Use drag-and-drop interface
• ✅ Centralized Control: Update views in Pylar, workflows get updates
• ✅ Secure Access: Through governed views
• ✅ Full Visibility: Monitor usage in Pylar Evals
• ✅ Open Source: Self-host or use cloud

​

Troubleshooting

​

Issue: Authentication fails

• Verify Bearer Token format: Bearer YOUR_TOKEN
• Check header name is exactly “Authorization”
• Ensure no extra spaces in token
• Test authentication separately

​

Issue: Tool not found

• Verify tool name matches exactly
• Check tools are published in Pylar
• Review JSON body structure
• Test tool in Pylar first

​

Issue: Request format error

• Verify JSON-RPC format is correct
• Check method name: tools/call
• Ensure parameters match tool schema
• Review n8n expression syntax

​

Best Practices

​

Use Expressions

{
  "event_type": "{{ $json.event_type }}",
  "start_date": "{{ $json.start_date }}"
}

​

Error Handling

1. Add IF node after HTTP Request
2. Check for errors in response
3. Route to error handling or retry logic

​

Data Transformation

1. Use Set node to extract specific fields
2. Transform data for next steps
3. Map to other workflow nodes

​

Next Steps

• Publishing Tools - Publish your tools
• Evals - Monitor n8n workflow usage
• Other Builders - Connect to other platforms