n8n-nodes-humanly

n8n community nodes for Humanly AI-powered image classification and validation. Classify images to detect their type (Vehicle, Property, Retail, General) or validate images against a specific type -- directly from your n8n workflows.

Installation

In n8n Desktop / Self-Hosted

1. Go to Settings > Community Nodes
2. Enter n8n-nodes-humanly
3. Click Install

Via npm (self-hosted)

cd ~/.n8n
npm install n8n-nodes-humanly

Restart n8n after installation.

Credentials Setup

Before using the Humanly nodes, configure your API credentials:

1. In n8n, go to Credentials > Add Credential
2. Search for Humanly API
3. Fill in the required fields:

Credentials are stored securely by n8n and used to authenticate all requests to the Humanly API.

Nodes

Humanly (Action Node)

The main Humanly node provides three operations for image analysis.

Classify Image

Upload an image and detect its type automatically.

Inputs:

• Binary Property -- the name of the binary property containing the image file (default: data)

Outputs:

• jobId -- the processing job ID to track the result

Example: Connect an HTTP Request node (to fetch an image) or a Read Binary File node, then connect the Humanly node with "Classify Image" selected.

Validate Image

Upload an image and check whether it matches a specific classification type.

Inputs:

• Binary Property -- the name of the binary property containing the image file (default: data)
• Classification Type -- the type to validate against: Vehicle, Property, Retail, or General

Outputs:

• jobId -- the processing job ID to track the result

Get Job

Retrieve the current status and results of a processing job.

Inputs:

• Job ID -- the job ID returned from a classify or validate operation

Outputs:

• id -- job identifier
• type -- classification or validation
• status -- pending, processing, completed, or failed
• createdAt, startedAt, completedAt -- timestamps
• data -- result data (when completed):
  • For classification: { detectedType, confidence }
  • For validation: { classificationType, result, probability, explanation }
• error -- error details (when failed): { code, message }

Humanly Trigger (Polling Trigger)

Polls a job until it reaches a terminal state (completed or failed), then triggers the workflow.

Inputs:

• Job ID -- the job ID to monitor
• Poll Interval (ms) -- how often to check the job status (default: 5000)
• Timeout (ms) -- maximum time to wait before giving up (default: 120000)

Outputs:

• The full job object (same fields as Get Job above)

Use this when you want a workflow to start automatically once an image analysis job finishes. Connect it to downstream nodes that process the classification or validation results.

Example Workflows

Classify an Image from URL

[HTTP Request] -> [Humanly: Classify Image] -> [Humanly: Get Job] -> [IF: status == completed] -> [Process Result]

1. HTTP Request -- download the image (stores as binary data)
2. Humanly: Classify Image -- sends the image for classification
3. Wait -- pause for processing (or use the Humanly Trigger instead)
4. Humanly: Get Job -- retrieve the result
5. IF -- check if status is completed
6. Process Result -- use the classification data downstream

Validate and Route by Result

[Webhook] -> [HTTP Request] -> [Humanly: Validate Image (Vehicle)] -> [Humanly Trigger] -> [Switch: result] -> [Pass / Fail branches]

1. Webhook -- receives a request with an image URL
2. HTTP Request -- downloads the image
3. Humanly: Validate Image -- validates against "Vehicle" type
4. Humanly Trigger -- waits for the job to complete
5. Switch -- routes based on data.result (pass or fail)

Error Handling

The node surfaces API errors as n8n execution errors with descriptive messages:

• Authentication errors -- check your Humanly API credentials
• Rate limit errors -- you've exceeded the API rate limit; retry after the indicated period
• Quota errors -- your usage quota has been exceeded
• API errors -- general server errors with status code and message

Compatibility

• n8n >= 1.0.0
• Node.js >= 20.0.0