Get the FREE Ultimate OpenClaw Setup Guide →

awesome-remote s

A curated list of Hosted & Managed Model Context Protocol (MCP) Servers accessible via a simple URL endpoint.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio sylviangth-awesome-remote-mcp-servers docker run -i awesome-remote-mcp-server-image \
  --env MCP_API_KEY="your-api-key-if-required" \
  --env MCP_SERVER_URL="https://awesome.remote/mcp"

How to use

Awesome Remote MCP Servers are hosted endpoints that expose a suite of AI-accessible tools via the MCP (Model Context Protocol). Each server URL acts as a gateway to a set of integrations, data sources, and actions that you can invoke from your AI agents. You don’t run code locally; instead you send requests to the remote MCP URL and receive structured responses that your agent can consume. The repository showcases a curated collection of these endpoints, along with details on the kinds of capabilities each server provides and how to authenticate if needed. When integrating, you typically pass input data, specify the desired tool or action, and handle the output within your agent’s reasoning loop. The tools span categories like data retrieval, automation, memory access, and third-party service integrations, depending on the server chosen.

How to install

Prerequisites:

  • Access to the Awesome Remote MCP Servers (you may need an API key or account with the hosting provider).
  • Basic knowledge of the MCP protocol and how to route requests from your AI agent.

Option A: Connect to a hosted MCP server directly (no local installation required)

  • Obtain the server URL and any required authentication token from the provider.
  • Use your agent to call the MCP endpoints using HTTP requests or an MCP client library.

Option B: Run a local proxy or client to interact with remote MCP servers

  • Ensure Docker is installed on your machine.
  • Use the following command to start a client container that can talk to the hosted MCP server:
# Replace the URL and API key with your values
docker run -it --rm \
  -e MCP_SERVER_URL="https://awesome.remote/mcp" \
  -e MCP_API_KEY="your-api-key-if-required" \
  awesome-remote-mcp-client:latest

Option C: If the MCP server is provided as a Node.js/JS package

  • Install Node.js (12+ or current LTS).
  • Install the package and run the server client as documented by the provider. Example (adjust package name as needed):
npm install -g awesome-remote-mcp-client
awesome-remote-mcp-client --server https://awesome.remote/mcp --token your-api-key

Note: Because this repository is a catalog of hosted endpoints, most users will simply route requests to the given server URLs rather than running local server code.

Additional notes

Tips and considerations:

  • Always verify the MCP server’s authentication requirements (API keys, OAuth tokens, or no auth at all).
  • Check the server’s documentation for available tools, input schemas, and output formats.
  • Implement retry logic and rate limiting in your agent when calling remote MCP servers to handle transient network issues.
  • Monitor latency and reliability; hosted servers may have different SLAs compared to self-hosted options.
  • If you switch servers, reuse your agent’s request templates but adjust the tool identifiers and input schemas to match the new endpoint.

Related MCP Servers

Sponsor this space

Reach thousands of developers