Wiz MCP

A command-line tool that connects to N8N MCP servers using smolagents and LiteLLM.

Usage

Basic Usage

uv run main.py --n8n-api-key YOUR_API_KEY "Your message here"

With Custom LLM Model

uv run main.py \
  --llm-model "ollama/llama3:latest" \
  --n8n-api-key YOUR_API_KEY \
  "Analyze my recent emails"

With Custom LLM URL

uv run main.py \
  --llm-url "http://localhost:8080" \
  --llm-model "gpt-4o-mini" \
  --n8n-api-key YOUR_API_KEY \
  "Help me organize my inbox"

With Custom N8N SSE URL

uv run main.py \
  --n8n-sse-url "https://your-n8n.domain.com/mcp/gmail-v2/sse" \
  --n8n-api-key YOUR_API_KEY \
  "What are my unread emails?"

Full Configuration

uv run main.py \
  --llm-model "ollama/gemma3:latest" \
  --llm-url "http://localhost:11434" \
  --n8n-sse-url "https://n8n.example.com/mcp/gmail-v2/sse" \
  --n8n-api-key YOUR_API_KEY \
  --temperature 0.2 \
  "Draft a response to my latest email from John"

Options

• --llm-model, -m: The model to use for the agent (default: ollama/gemma3:latest)
• --llm-url, -u: The base URL for the LLM API (default: http://localhost:11434)
• --n8n-sse-url, -s: The SSE URL for the MCP server (default: https://n8n.example.com/mcp/gmail-v2/sse)
• --n8n-api-key, -k: The API key for N8N (required)
• --temperature, -t: The temperature for the LLM model (default: 0.1)

Environment Variables

You can also set the N8N API key as an environment variable:

export N8N_API_KEY="your-api-key-here"
uv run main.py --n8n-api-key "$N8N_API_KEY" "Your message"