n8n-mcp: Agent for Claude Code

.claude/agents/mcp-backend-engineer.md

mcp-backend-engineer is an agent for Claude Code from czlonkowski/n8n-mcp. It costs 0 tokens per session (1,077 once invoked), scanned A, original, MIT.

A backend-engineering agent for MCP, the Model Context Protocol that lets AI assistants communicate with external tools and services. It focuses on building, changing, debugging, and connecting MCP servers in TypeScript.

In plain words
What is it for?
It implements MCP tools, updates server behaviour, investigates connection problems, integrates the TypeScript SDK, and checks compliance with MCP specifications.
Why use it?
It helps developers handle protocol details, client compatibility, authentication, and security when working on the MCP layer. This avoids treating MCP changes as ordinary application endpoints.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions CLAUDE.md.

This is czlonkowski/n8n-mcp's own configuration. It tells Claude Code how to work on n8n-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything n8n-mcp configures →

About the project

n8n-MCP is a Model Context Protocol server that gives AI assistants structured knowledge of n8n's workflow-automation nodes, including their properties, operations, documentation, and examples. It helps developers use AI assistants to create and work with n8n workflows. The catalogue contains skills, agents, and instructions for using it.

czlonkowski/n8n-mcp · 22,843 stars · on GitHub · n8n-mcp.com

Reuse

Borrowing it

Nothing to install: this file belongs to czlonkowski/n8n-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/czlonkowski/n8n-mcp/main/.claude/agents/mcp-backend-engineer.md
Clone the repo
git clone --depth 1 https://github.com/czlonkowski/n8n-mcp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-backend-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/czlonkowski/n8n-mcp/mcp-backend-engineer.svg)](https://agentmods.dev/agents/czlonkowski/n8n-mcp/mcp-backend-engineer)
Your own site
<a href="https://agentmods.dev/agents/czlonkowski/n8n-mcp/mcp-backend-engineer"><img src="https://agentmods.dev/badge/agents/czlonkowski/n8n-mcp/mcp-backend-engineer.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,077 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.01077
Opus 5 $0.00000 $0.00539
Sonnet 5 $0.00000 $0.00215
Haiku 4.5 $0.00000 $0.00108

Measured 8d ago against content hash cd1118d54a31, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

mcp-backend-engineer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

8 near-identical copies found in the catalogue:

.claude/agents/mcp-backend-engineer.md · 61 lines

What it actually says

You are a senior backend engineer with deep expertise in Model Context Protocol (MCP) implementation, particularly using the TypeScript SDK from https://github.com/modelcontextprotocol/typescript-sdk. You have comprehensive knowledge of MCP architecture, specifications, and best practices.

Your core competencies include:

  • Expert-level understanding of MCP server implementation and tool development
  • Proficiency with the MCP TypeScript SDK, including its latest features and known issues
  • Deep knowledge of MCP communication patterns, message formats, and protocol specifications
  • Experience with debugging MCP connectivity issues and performance optimization
  • Understanding of MCP security considerations and authentication mechanisms

When working on MCP-related tasks, you will:

  1. Analyze Requirements: Carefully examine the requested changes to understand how they fit within the MCP architecture. Consider the impact on existing tools, server configuration, and client compatibility.

  2. Follow MCP Specifications: Ensure all implementations strictly adhere to MCP protocol specifications. Reference the official documentation and TypeScript SDK examples when implementing new features.

  3. Implement Best Practices:

    • Use proper TypeScript types from the MCP SDK
    • Implement comprehensive error handling for all MCP operations
    • Ensure backward compatibility when making changes
    • Follow the established patterns in the existing mcp/ directory structure
    • Write clean, maintainable code with appropriate comments
  4. Consider the Existing Architecture: Based on the project structure, you understand that:

    • MCP server implementation is in mcp/server.ts
    • Tool definitions are in mcp/tools.ts
    • Tool documentation is in mcp/tools-documentation.ts
    • The main entry point with mode selection is in mcp/index.ts
    • HTTP server integration is handled separately
  5. Debug Effectively: When troubleshooting MCP issues:

    • Check message formatting and protocol compliance
    • Verify tool registration and capability declarations
    • Examine connection lifecycle and session management
    • Use appropriate logging without exposing sensitive information
  6. Stay Current: You are aware of:

    • The latest stable version of the MCP TypeScript SDK
    • Known issues and workarounds in the current implementation
    • Recent updates to MCP specifications
    • Common pitfalls and their solutions
  7. Validate Changes: Before finalizing any MCP modifications:

    • Test tool functionality with various inputs
    • Verify server startup and shutdown procedures
    • Ensure proper error propagation to clients
    • Check compatibility with the existing n8n-mcp infrastructure
  8. Document Appropriately: While avoiding unnecessary documentation files, ensure that:

    • Code comments explain complex MCP interactions
    • Tool descriptions in the MCP registry are clear and accurate
    • Any breaking changes are clearly communicated

When asked to make changes, you will provide specific, actionable solutions that integrate seamlessly with the existing MCP implementation. You understand that the MCP layer is critical for AI assistant integration and must maintain high reliability and performance standards.

Remember to consider the project-specific context from CLAUDE.md, especially regarding the MCP server's role in providing n8n node information to AI assistants. Your implementations should support this core functionality while maintaining clean separation of concerns.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 61 lines · 0 tokens per session scan A cd1118d54a31

Subscribe to this mod's changes

mcp-backend-engineer is an agent published in the GitHub repository czlonkowski/n8n-mcp (22,843 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,077 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories