n8n-workflow

name	n8n-workflow
description	Design and build n8n automation workflows. Use when creating automations, connecting APIs, optimizing workflows, or troubleshooting n8n implementations.
version	1.0.0
license	MIT

n8n Workflow Specialist

Note: Review PROFILE.md in this skill folder for user-specific use cases, common workflows, and integration preferences.

Master Briefing: Global brand voice at ~/.superskills/master-briefing.yaml applies automatically. Skill profile overrides when conflicts exist.

Design, build, and optimize n8n automation workflows that connect tools, automate processes, and eliminate repetitive tasks.

Core Workflow

1. Requirements & Planning

• Understand automation goal and trigger conditions
• Identify systems and data to connect
• Map current manual process
• Define success criteria and error handling

2. Workflow Design

• Choose appropriate trigger (webhook, schedule, manual)
• Plan node sequence and logic flow
• Design data transformation steps
• Plan error handling and fallbacks

3. Node Configuration

• Configure each node with proper credentials
• Set up data mapping between nodes
• Implement conditional logic and routing
• Add error handling nodes

4. Testing & Optimization

• Test with real data
• Verify error handling works
• Optimize for performance
• Document workflow purpose and maintenance

Quality Checklist

• Clear trigger conditions defined
• Error handling implemented
• Credentials properly configured
• Data validation at key steps
• Workflow documented with notes
• Tested with real data
• Performance optimized (minimize API calls)
• Monitoring/alerting configured

n8n Fundamentals

See PROFILE.md for:

• Common workflow templates
• Integration preferences
• Use case examples

Core Concepts:

Nodes:

• Trigger Nodes: Start workflow (webhook, schedule, manual)
• Regular Nodes: Process data, call APIs, transform
• Logic Nodes: IF, Switch, Merge, Split
• Output Nodes: Send data to destinations

Data Flow:

• Each node receives input from previous node
• Nodes output data for next node
• Use expressions to reference previous node data: {{ $json.fieldName }}
• Multiple items can flow through workflow

Credentials:

• Store in n8n credentials manager
• Never hardcode API keys
• Use environment variables for sensitive data

Common Node Types

Triggers

• Webhook: Receive HTTP requests
• Schedule: Cron-based timing
• Manual Trigger: Button click
• Email Trigger: IMAP monitoring
• RSS Feed: Monitor feed for new items

Data Processing

• Set: Add/modify data fields
• Code: JavaScript for custom logic
• Function: Transform data
• Split In Batches: Process large datasets
• Merge: Combine multiple inputs

Integrations

• HTTP Request: Call any API
• OpenAI: GPT completions, embeddings
• Anthropic: Claude API
• Airtable: Database operations
• Google Sheets: Spreadsheet operations
• Email: Send via SMTP
• Slack: Post messages
• Notion: Database queries

Logic & Control

• IF: Conditional branching
• Switch: Multiple conditions
• Filter: Remove items
• Wait: Delay execution
• Error Trigger: Catch workflow errors

Workflow Templates

API to Spreadsheet Logger

Trigger: Webhook
→ HTTP Request (fetch data from API)
→ Set (format data)
→ Google Sheets (append row)
→ Slack (send notification)

Content Distribution Pipeline

Trigger: Manual or Webhook
→ OpenAI (generate variations)
→ IF (check content length)
  → True: Split content
  → False: Continue
→ Multiple paths:
  → LinkedIn (post)
  → Twitter (post)
  → Email (draft)
→ Airtable (log results)

Lead Enrichment Workflow

Trigger: Webhook (new lead)
→ HTTP Request (enrich from Clearbit)
→ IF (score > threshold)
  → True: 
    → Slack (notify sales)
    → CRM (update record)
  → False:
    → Email (nurture sequence)

Research Automation

Trigger: Schedule (daily)
→ RSS (fetch news)
→ OpenAI (summarize articles)
→ Filter (relevance check)
→ Notion (create page)
→ Email (send digest)

Best Practices

Error Handling

• Add Error Trigger node to catch failures
• Use IF nodes to validate data before processing
• Set Retry on Fail for API calls
• Send error notifications to Slack/email

Performance

• Minimize API calls (batch when possible)
• Use Split In Batches for large datasets
• Cache frequently used data
• Avoid unnecessary data transformations

Maintenance

• Use Sticky Notes to document complex logic
• Name nodes descriptively
• Version control workflow exports
• Test regularly with production data
• Monitor execution logs

Security

• Never hardcode credentials
• Use environment variables
• Validate webhook signatures
• Sanitize user inputs
• Limit webhook exposure

Debugging Tips

Common Issues:

"No data" error:

• Check previous node executed successfully
• Verify data path: {{ $json.fieldName }}
• Ensure node received input

"Missing credentials" error:

• Add credentials in Settings
• Select correct credential in node
• Test credential connection

"Rate limit" error:

• Add Wait node between requests
• Implement exponential backoff
• Use Split In Batches with delay

"Workflow timeout" error:

• Split into smaller workflows
• Use Split In Batches
• Optimize slow nodes

Expression Examples

// Access previous node data
{{ $json.email }}
{{ $json.user.name }}

// Current date/time
{{ $now.format('YYYY-MM-DD') }}
{{ $today.toISO() }}

// String manipulation
{{ $json.name.toLowerCase() }}
{{ $json.text.substring(0, 100) }}

// Conditional logic
{{ $json.score > 80 ? 'high' : 'low' }}

// Array operations
{{ $json.items.length }}
{{ $json.tags.join(', ') }}

// Math
{{ $json.price * 1.2 }}
{{ Math.round($json.value) }}

Avoid

• No Error Handling: Assume success → Add error triggers and validation
• Hardcoded Values: API keys in nodes → Use credentials manager
• No Testing: Deploy untested → Test with real data first
• Complex Single Workflow: 50+ nodes → Break into sub-workflows
• No Documentation: Unclear purpose → Add sticky notes, descriptive names
• Ignore Logs: Don't check executions → Monitor for failures

Escalate When

• Need custom node development
• Complex authentication (OAuth2, custom)
• Performance issues at scale
• Webhook security concerns
• Integration not available as node
• Workflow too complex (consider custom code)