N8nOps - Autonomous n8n Workflow Automation Agent

An expert-level AI agent that lives inside your n8n instance, capable of creating, debugging, testing, monitoring, and optimizing workflows via the n8n REST API. Built on the OpenClaw agent framework.

What It Does

N8nOps acts as your personal automation architect. Instead of manually building workflows in the n8n UI, you describe what you need in plain language, and the agent handles the rest:

• Create workflows from natural language descriptions
• Debug & fix failed executions by tracing errors to specific nodes
• Monitor health with automatic heartbeat checks every 30 minutes
• Manage credentials and guide setup when needed
• Build AI pipelines with LangChain integration (RAG, agents, tools, memory)
• Optimize existing workflows for performance and reliability

Tech Stack

Component	Technology
Primary LLM	Anthropic Claude Sonnet 4
Fallback LLM	OpenAI GPT-4o
Target Platform	n8n 2.0.0
Agent Framework	OpenClaw
Communication	REST API via curl

Project Structure

openclaw-n8n-agent/
├── openclaw.json          # Agent configuration
├── SOUL.md                # Core identity & n8n expertise
├── IDENTITY.md            # Personality definition
├── AGENTS.md              # Workspace protocols
├── TOOLS.md               # n8n connection details & API patterns
├── BOOTSTRAP.md           # First-run setup guide
├── MEMORY.md              # Long-term memory
├── HEARTBEAT.md           # Health monitoring protocol
├── USER.md                # User profile
├── memory/                # Session-based memory files
└── skills/n8n/            # Expertise modules
    ├── SKILL.md            # n8n mastery overview
    ├── node-catalog.md     # 65+ verified node types
    ├── ai-patterns.md      # AI workflow connection patterns
    ├── api-reference.md    # n8n REST API endpoints
    └── templates.md        # Ready-to-use workflow templates

Features

Workflow Lifecycle Management

• Create, edit, activate/deactivate, and delete workflows
• Validate workflow JSON before deployment
• Show diffs when updating workflows for human review

Execution Monitoring

• List and filter executions by status, workflow, or date range
• Diagnose failed executions and identify root causes
• Retry failed workflows and clean up old logs

AI/LangChain Integration

• Build AI agent pipelines with tool use
• Configure LLM models (OpenAI, Anthropic, Ollama, Google Gemini)
• Set up conversational memory (Buffer Window, Redis, Postgres)
• Implement RAG workflows with vector store integration
• Supported vector stores: Pinecone, Postgres, Qdrant, In-Memory

Health Monitoring

• Automatic heartbeat checks every 30 minutes
• Alerts on failed executions and critical workflow failures
• Tracks active workflow count and instance health

Getting Started

Prerequisites

• n8n instance running (default: http://localhost:5678)
• n8n version 2.0.0
• n8n API key

Setup

1. Verify n8n is accessible:

   curl -s -o /dev/null -w "%{http_code}" http://localhost:5678/healthz
   # Expected: 200

2. Create an API key in n8n UI:

   • Go to Settings > API > Create API Key

3. Test the API key:

   curl -s -H "X-N8N-API-KEY: YOUR_KEY" http://localhost:5678/api/v1/workflows

4. Set environment variable:

   export N8N_API_KEY=YOUR_API_KEY

5. Run bootstrap to inventory workflows and credentials on first launch.

Configuring n8n for OpenClaw

1. Install & Run n8n

You need a running n8n 2.0.0 instance. You can set it up with Docker or npm:

Docker (recommended):

docker run -it --rm \
  --name n8n \
  -p 5678:5678 \
  -v n8n_data:/home/node/.n8n \
  n8nio/n8n:2.0.0

npm:

npx n8n@2.0.0

n8n will be available at http://localhost:5678.

2. Enable the n8n API

1. Open http://localhost:5678 in your browser
2. Complete the initial setup (create owner account)
3. Go to Settings > API
4. Click Create API Key
5. Copy the generated key

3. Configure OpenClaw Connection

Edit openclaw.json to point to your n8n instance:

{
  "agent": {
    "model": "anthropic/claude-sonnet-4-20250514",
    "fallbackModel": "openai/gpt-4o",
    "workspace": "./",
    "thinkingDefault": "high",
    "timeoutSeconds": 600,
    "heartbeat": {
      "every": "30m"
    }
  },
  "tools": {
    "exec": { "enabled": true, "policy": "standard" },
    "web_fetch": { "enabled": true },
    "web_search": { "enabled": true },
    "browser": { "enabled": true },
    "apply_patch": { "enabled": true },
    "diffs": { "enabled": true }
  },
  "env": {
    "N8N_BASE_URL": "http://localhost:5678",
    "N8N_API_BASE": "http://localhost:5678/api/v1"
  }
}

If your n8n runs on a different host/port, update the env.N8N_BASE_URL and env.N8N_API_BASE values accordingly.

4. Set Your API Key

Export the n8n API key as an environment variable:

export N8N_API_KEY=your-api-key-here

5. Verify the Connection

curl -s -H "X-N8N-API-KEY: $N8N_API_KEY" http://localhost:5678/api/v1/workflows

If you get a JSON response with workflow data (or an empty data array), the connection is working.

6. Bootstrap the Agent

On first run, the agent will:

1. Verify n8n instance accessibility
2. Inventory all existing workflows and credentials
3. Populate MEMORY.md with the workflow catalog
4. Update TOOLS.md with credential inventory
5. Load skill files from skills/n8n/ for node types, AI patterns, and templates

After bootstrap completes, the agent is ready to build, debug, and manage workflows.

Key Configuration Options

Option	Default	Description
agent.model	anthropic/claude-sonnet-4-20250514	Primary LLM for reasoning
agent.fallbackModel	openai/gpt-4o	Fallback when primary is unavailable
agent.thinkingDefault	high	Reasoning depth (high = deep analysis)
agent.timeoutSeconds	600	Max seconds per operation
agent.heartbeat.every	30m	Health check interval
session.reset.idleMinutes	10080	Session reset after 7 days idle
env.N8N_BASE_URL	http://localhost:5678	n8n instance URL
env.N8N_API_BASE	http://localhost:5678/api/v1	n8n API endpoint

n8n API Endpoints Used

The agent communicates with n8n through these REST API endpoints:

Method	Endpoint	Purpose
GET	/api/v1/workflows	List all workflows
POST	/api/v1/workflows	Create a new workflow
GET	/api/v1/workflows/:id	Get workflow details
PUT	/api/v1/workflows/:id	Update a workflow
DELETE	/api/v1/workflows/:id	Delete a workflow
POST	/api/v1/workflows/:id/activate	Activate a workflow
POST	/api/v1/workflows/:id/deactivate	Deactivate a workflow
POST	/api/v1/workflows/:id/run	Trigger execution
GET	/api/v1/executions	List executions
GET	/api/v1/executions/:id	Get execution details
GET	/api/v1/credentials	List credentials

All API calls require the X-N8N-API-KEY header for authentication.

Safety

• Confirms before activating/deactivating production workflows
• Confirms before deleting workflows
• Never hardcodes API keys or secrets
• Always reads current state before modifying (GET before PUT)
• Shows diffs for all workflow updates

Security Verification

This repository has been scanned and verified as safe by VirusTotal — a trusted security service that aggregates results from 95+ antivirus engines and URL scanners.

Scan Result: ✅ 0/95 — No security vendors flagged this repository as malicious.

🔗 View full VirusTotal report

Detail	Value
Scanned URL	https://github.com/samansalari/Openclaw-N8N
Detection Ratio	0 / 95
Community Score	+1 👍
Status	200 OK
Content Type	text/html; charset=utf-8

All 95 security vendors — including BitDefender, CyRadar, AlienVault, CRDF, Dr.Web, and others — returned a Clean verdict.

Author

Saman Salari - Developer, Product Designer & AI Enthusiast

Based in the UK, focused on AI engineering to create smart, impactful digital experiences. 10+ years of experience building scalable web applications and automation systems using React, Next.js, Node.js, and AI-powered development tools.

• Website: samansalari.com
• LinkedIn: linkedin.com/in/samansalari
• X: x.com/samansalari
• Email: me@samansalari.com

License

This project is open source. See LICENSE for details.