n8n-nodes-everyrow

This is an n8n community node for Everyrow - AI-powered data operations for your workflows.

Everyrow enables you to perform intelligent data operations using AI, including ranking, deduplication, merging, screening, and research tasks.

n8n is a fair-code licensed workflow automation platform.

Installation

For n8n Users (Community Node)

Follow the installation guide in the n8n community nodes documentation.

npm install n8n-nodes-everyrow

For n8n Cloud Users

Once this node is verified, it will be available directly in n8n Cloud. Until then, you can use our HTTP Request workflow templates - see templates/n8n-cloud-or-no-extension/ for ready-to-import workflows that work on any n8n instance without installing custom nodes.

Local Development Setup

To test this node locally during development:

Prerequisites

1. Node.js 18+ - Install via nvm or nodejs.org
2. pnpm - Install with npm install -g pnpm
3. n8n - Install globally with npm install -g n8n
4. Everyrow API Key - Get one at everyrow.io/settings/api-keys

Step 1: Clone and Build

git clone https://github.com/futuresearch/n8n-nodes-everyrow.git
cd n8n-nodes-everyrow

# Install dependencies
pnpm install

# Build the node
pnpm build

Step 2: Install in n8n

# Create n8n custom nodes directory
mkdir -p ~/.n8n/nodes
cd ~/.n8n/nodes

# Install the local package
npm init -y
npm install /path/to/n8n-nodes-everyrow

Step 3: Start n8n

n8n start

Open http://localhost:5678 in your browser.

Step 4: First-Time n8n Setup

1. Create an account - n8n will prompt you to create a local account (email + password)
2. Skip or complete onboarding - You can skip the onboarding wizard

Step 5: Configure Everyrow Credentials

1. Go to Credentials (left sidebar) → Add Credential
2. Search for "Everyrow"
3. Enter your API key from everyrow.io/settings/api-keys
4. Click Save

Step 6: Create a Test Workflow

Import a template from templates/self-hosted-with-extension/ or build manually:

1. Click Add Workflow
2. Add a Manual Trigger node
3. Add a Code node with sample data:

   const companies = [
     { name: "OpenAI", description: "AI research company" },
     { name: "Stripe", description: "Payment processing" },
     { name: "Anthropic", description: "AI safety company" }
   ];
   return companies.map(c => ({ json: c }));

4. Add the Everyrow node:
   • Select Data Operations → Rank
   • Task: "Score by AI relevance, 0-100"
   • Field Name: score
5. Add a Wait node (5 seconds)
6. Add another Everyrow node:
   • Select Task → Get Status
   • Task ID: {{ $('Everyrow').item.json.task_id }}
7. Add an If node to check {{ $json.status }} equals "completed"
8. On true branch, add Everyrow → Task → Get Result
9. On false branch, loop back to the Wait node
10. Connect the nodes and click Test Workflow

Running Tests

Tests require a valid API key. Create a .env file (gitignored):

cp .env.example .env
# Edit .env and add your API key

Then run:

pnpm test         # Run tests once
pnpm test:watch   # Run tests in watch mode

Development Commands

pnpm build        # Build the node
pnpm dev          # Watch mode for development
pnpm lint         # Run linter
pnpm test         # Run tests (requires .env with API key)

Troubleshooting

Node not appearing in n8n?

• Ensure the package is built (pnpm build)
• Check n8n logs for loading errors: N8N_LOG_LEVEL=debug n8n start
• Verify the symlink exists: ls -la ~/.n8n/nodes/node_modules/

"Everyrow" credential type not found?

• Restart n8n after installing the package
• Check that dist/credentials/EveryrowApi.credentials.js exists

API errors?

• Verify your API key is correct
• Check the API URL (should be: https://engine.futuresearch.ai/api/v0)

Operations

Data Operations

| Operation | Description | |-----------|-------------| | Rank | Score and rank rows based on AI-evaluated criteria. Useful for prioritizing leads, scoring content relevance, or ranking search results. | | Dedupe | Remove duplicate rows using AI matching. Handles fuzzy matching, name variations, and semantic similarity. | | Screen | Filter rows based on complex criteria that require AI understanding. Filter companies by funding stage, filter products by features, etc. | | Merge | Join two tables using AI-powered matching. Merge customer lists, match products across databases, etc. |

Agent Operations

| Operation | Description | |-----------|-------------| | Agent Map | Run an AI research agent on each row to enrich data with web research, analysis, or complex reasoning. |

Task Operations

| Operation | Description | |-----------|-------------| | Get Status | Check the status of a submitted task (pending, running, completed, failed). | | Get Result | Retrieve the results of a completed task. |

Note: All operations return a task_id immediately. Use the Task operations with a polling loop to wait for completion and retrieve results. See the workflow templates for examples.

Credentials

To use this node, you need an Everyrow API key:

1. Sign up at everyrow.io
2. Go to Settings > API Keys
3. Create a new API key
4. Add the credentials in n8n

Example Workflows

Ready-to-import workflow templates are available in the templates/ directory.

Rank Companies by AI Relevance

[Data Source] → [Everyrow: Rank] → [Wait 5s] → [Everyrow: Get Status] → [If Completed?]
                                                                              ↓ Yes
                                                                    [Everyrow: Get Result]
                                                                              ↓ No
                                                                    [Loop back to Wait]

1. Add a data source node (Google Sheets, Airtable, etc.)
2. Add the Everyrow node with Data Operations → Rank
3. Set the task: "Score each company by their relevance to enterprise AI infrastructure"
4. Add a Wait node, then poll with Task → Get Status
5. Use an If node to check if status == "completed"
6. Get results with Task → Get Result

Deduplicate a Contact List

1. Import your contact list
2. Add the Everyrow node with Data Operations → Dedupe
3. Set the equivalence relation: "Two contacts are duplicates if they represent the same person, even if names are spelled differently or companies have changed"
4. Add the polling loop (Wait → Get Status → If → Get Result)
5. Output the deduplicated list

Research and Enrich Data

1. Add your data source
2. Add the Everyrow node with Agent Operations → Agent Map
3. Set the task: "Research each company and find their latest funding round, founding year, and key products"
4. Define the response schema with the fields you want
5. Add the polling loop to wait for completion
6. Get enriched data with AI-researched information

Workflow Pattern

Everyrow operations are asynchronous. The recommended workflow pattern is:

1. Submit Operation - Returns a task_id immediately
2. Wait - Use n8n's Wait node (e.g., 5-10 seconds)
3. Check Status - Use Task → Get Status to poll
4. Loop or Continue - If not completed, loop back to Wait; if completed, continue
5. Get Results - Use Task → Get Result to retrieve data

See templates/self-hosted-with-extension/ for complete workflow examples.

Configuration Options

• Session Name: Name for the Everyrow session (visible in dashboard)
• Task ID: Reference to a previously submitted task (for Get Status / Get Result)

Each operation has specific configuration options - see the node UI for details.

Resources

• Everyrow Documentation
• n8n Community Nodes Documentation
• Report Issues

License

MIT