awesome s

A collection of MCP servers

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio pipedreamhq-awesome-mcp-servers node path/to/server.js \
  --env MCP_HOST="Host for the MCP server (default: 0.0.0.0)" \
  --env MCP_PORT="Port for the MCP server to listen on (default: 8080)"

How to use

The Awesome MCP Servers page is a curated directory of MCP server implementations that extend AI capabilities with real-world contextual services. This MCP ecosystem includes servers that provide access to files, databases, APIs, and other resources to enable AI agents to interact more intelligently with your environment. To get started, pick a server from the catalog that matches your use case (for example, a file storage MCP server for accessing documents or a database MCP server for querying data) and run it using the recommended deployment method. Once running, you can connect your MCP-enabled AI agent to the server's endpoints to request contextual data, run actions, or perform operations that the server supports. Each server typically exposes a set of APIs or endpoints designed to be consumed by MCP-enabled agents, so consult the server's docs for exact capabilities and authentication details.

When leveraging these servers, you can chain multiple MCP servers to build richer workflows. For example, an agent could fetch data from a database MCP server, enrich it with AI processing from an AI/ML API MCP server, and then store results back into a storage MCP server. The catalog also includes developer-oriented tools and experimental projects, so you can experiment with new integrations and broaden your model’s operational context.

How to install

Prerequisites:

Node.js and npm (for servers published as Node-based samples) or Python (for Python-based samples) depending on the deployment method you choose.
Basic command-line familiarity and network access to reach required endpoints.

Installation steps (generalized):

Choose a server from the catalog that matches your use case (e.g., file storage, databases, or API integrations).
Follow the server's preferred deployment method. Examples:
- Node-based server (if applicable):
  - git clone <server-repo-url>
  - cd <server-repo>
  - npm install
  - npm run start
- npx-based (if the server is distributed as an npm package):
  - npx -y <package-name>
- Python/uv (if the server is provided as a Python package with uvicorn/uv):
  - pipx install <package-name> (or pip install <package-name> and run as instructed)
  - uvx run <module-or-script>
- Docker (if a container image is provided):
  - docker run -i <image-name>
Configure environment variables as required by the chosen server (see the server's docs for details).
Verify the server is reachable and note the authentication method and endpoints.
Connect your MCP-enabled agent to the server using the documented endpoint format and credentials.

Additional notes

Tips and common issues:

Ensure network access and firewall rules permit the MCP server to communicate with any external services it needs (databases, APIs, storage).
Some servers require API keys or OAuth tokens; keep credentials secure and rotate regularly.
If using Docker, ensure you expose the correct ports and mount any needed configuration files.
Check the server's documentation for supported MCP endpoints, request formats, and rate limits.
Environment variables often control ports, hostnames, authentication, and feature toggles; document and preserve them in your deployment environment.
When chaining servers, verify the latency and error handling across the chain to avoid cascading failures.