M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
N

n8n MCP Client Node

Community
Connect and call MCP servers in n8n workflows.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #355 of 438
Stars
★ 3.0k Popular
Transport
stdio (local process) · SSE · Streamable HTTP
Runtime
Node.js · Docker
Credentials
Optional API key
License
MIT
Last commit
⚠
Tools
6
40FMRS · D

This node enables seamless integration of MCP servers into n8n workflows, supporting multiple transports and use as a tool in AI agents. However, it requires proper security configuration, especially when using local commands.

Strongest · Setup experience 11/20 Weakest · Reliability 5/20

Reliability
5/20
Security and permissions
6/20
Maintenance
7/20
Documentation
11/20
Setup experience
11/20
Why each score
Reliability 5/20
The provided evidence is only the README; there is no package manifest, source, tests, or CI. This repository is an n8n client node rather than a standalone MCP server, so the 'init + tool-list handshake' cannot be observed in the review corpus. README claims STDIO, SSE, and HTTP Streamable transports and six operations, which makes a happy path plausible, but nothing verifies startup, error handling, or dependency control. Deductions for missing manifest, tests, and execution evidence; score kept low per static rules.
Security and permissions 6/20
No malware or token leakage is visible from the README; credentials are described as stored by n8n and can be passed via environment variables or headers. However, no source code verifies encryption, least-privilege scoping, output sanitization, or confirmation for executing remote tools that may have side effects. Data boundaries are unclear because the node deliberately sends data to external MCP servers. Deduction for unverifiable permission/confirmation handling; no red-line trigger.
Maintenance 7/20
License is MIT and the repository is not archived, with a public issue tracker, which are positive maintenance signals. However, the provided material contains no release history, commit cadence, dependency update health, contributor governance, or security-response channel. No explicit security policy file was supplied. Score reflects visible license and activity but missing governance and update evidence.
Documentation 11/20
The README is genuinely helpful: installation link, three transport credential formats, env var usage, operations list, AI-agent tool setup, Docker examples, and compatibility notes. Deductions: no troubleshooting section, no limits/cost disclosure, tool parameters are dynamic and only generically described, and the compatibility section omits HTTP Streamable while the top of the README deprecates SSE. There are also minor claims (e.g., MseeP badges) that cannot be verified from the supplied corpus.
Setup experience 11/20
Setup is well explained for a single platform: install as an n8n community node, create one of the three credential types, and configure operations; the Docker env var and mandatory N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE are documented. Still, it depends on the n8n runtime, requires manual environment configuration, and there is no one-command or verified client example in the source. Static calibration prevents awarding above 15 because no committed tests/CI are available to prove the setup path.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessRuns commands or codeUses the network

Best for

  • n8n users who want to integrate MCP tools into their automation flows.
  • Developers building AI agents that need to call external tools and data sources.
  • Teams that need to quickly connect to various MCP servers, whether local (via npx) or remote (HTTP).

Not for

  • Developers who need a direct programming interface rather than a graphical workflow.
  • Users without an n8n environment who just want a standalone MCP client.
  • Users who rely on SSE transport and cannot upgrade to HTTP Streamable (SSE is deprecated).

Required permissions

  • Requires the ability to install community nodes in n8n.
  • Requires setting the environment variable N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true to use the node as a tool.
  • With STDIO transport, it can execute arbitrary local commands (as defined by the user's command setting), potentially with system permissions.
  • Can make network requests to configured MCP servers, including sending additional headers.

Risks and side effects

  • STDIO commands can execute arbitrary local commands; if misconfigured (e.g., using untrusted servers), it may compromise the system.
  • SSE transport is deprecated and no longer updated, potentially having security vulnerabilities.
  • Environment variables (like API keys) may be stored in n8n credentials; if the n8n instance is not secure, there is a risk of exposure.
  • Running third-party MCP servers (e.g., downloaded via npx) may introduce insecure dependencies.

Setup

Before you start

Runtime：Node.js · Docker

MCP_BRAVE_API_KEY optionalsecret Brave API key passed to MCP servers in Docker deployments via the MCP_ prefix; obtain from Brave.
BRAVE_API_KEY optionalsecret Brave Search MCP server API key set via the credentials UI; obtain from Brave.
Other optional settings (1)
N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE optional Must be true to use this community node as a tool in n8n AI Agents; set via shell, Docker, or .env file.

To install the community node in n8n:

  1. Go to n8n's 'Settings' -> 'Community Nodes'.
  2. Enter the package name n8n-nodes-mcp-client and install.
  3. Alternatively, refer to the official n8n documentation for installing community nodes.

If you want to use this node as a tool in AI agents, you must set the environment variable N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true and restart n8n.

Check that it works

Add an MCP Client node in n8n and run the List Tools operation; if the server's tools (e.g. brave_search) are returned, the connection works.

Troubleshooting

  1. Ensure n8n version is at least 1.0.0.
  2. If you need to use tools in AI agents, confirm that N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true is set and n8n is restarted.
  3. Check that the command and arguments are correct, especially for local npx servers.
  4. Verify that the MCP server URL is reachable and that additional headers are properly set.
  5. If SSE connections fail, consider switching to HTTP Streamable (recommended).

Things to try

Once connected, you can ask your AI assistant things like:

  • List all available tools on this MCP server
  • Use the brave_search tool to find the latest AI news
  • Read the resource at the given URI
  • Show me the list of available prompts on the server

Tools 6

execute_tool writes
Execute a specific tool with given parameters.
get_prompt read-only
Retrieve a specific prompt template.
list_prompts read-only
List all available prompts.
list_resources read-only
List all resources provided by the server.
list_tools read-only
List all available tools with their descriptions and parameter schemas.
read_resource read-only
Read a specific resource by URI.

Use cases

Connect to external MCP servers like Brave Search, OpenAI, Serper, Weather, etc., within n8n workflows.
Let an AI agent call multiple tools via MCP, e.g., search for popular destinations, check weather, and find travel restriction news.
Manage MCP resources, prompts, and tools from within n8n.

Supported clients

Claude DesktopPartial support

Listed from the project's documentation, not tested by this site.

Overview

n8n-nodes-mcp-client is an n8n community node that lets you interact with Model Context Protocol (MCP) servers in your n8n workflows. It supports accessing resources, executing tools, and using prompts, and can be used as a tool in AI agents. It supports STDIO, SSE (deprecated), and HTTP Streamable transports.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 6ee215b30e51 Data synced 2026-10-12 Read the FMRS scoring method