n8n-skills: expert Claude Code skills for building and deploying production-ready n8n workflows

What it solves

Building n8n workflows programmatically via AI can be difficult due to complex expression syntax, incorrect tool usage, validation error loops, and misconfigured nodes. This project provides a set of specialized "skills" that teach AI assistants (specifically Claude) how to build production-ready n8n workflows and deploy self-hosted instances correctly.

How it works

It consists of 14 complementary skills and an enforcement layer designed for the n8n-mcp MCP server. These skills act as expert guides that activate automatically based on the user's query. They cover specific domains such as JavaScript/Python code nodes, AI agent design, binary data handling, and error recovery. An "enforcement layer" using hooks and a router skill proactively nudges the AI toward the correct guidance before and after tool use to prevent common mistakes.

Who it’s for

Developers and automation engineers using Claude Code, Claude.ai, or the Claude API who want to automate the creation, validation, and deployment of n8n workflows.

Highlights

• 14 Specialized Skills: Covers everything from expression syntax and node configuration to AI agent design and sub-workflows.
• Proactive Enforcement: Uses a router skill and pre/post-tool hooks to ensure the AI uses the right patterns at the right time.
• Code Node Expertise: Provides specific guidance for JavaScript and Python, including critical "gotchas" like webhook data access.
• Deployment Automation: Includes a dedicated skill for deploying production self-hosted n8n instances using Docker Compose and Caddy.
• Validation Support: Includes an expert skill to interpret validation errors and break error loops.