Skip to main content
Glama
deenesjakoptruzh
slackwarecps

MCP Simple Server

by slackwarecps
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Remote

MCP Simple Server (HTTP Bridge)

This is a simplified MCP (Model Context Protocol) server that exposes tools via an HTTP POST interface. It allows remote LLMs and agents to access local or specific functionality securely via a tunnel or direct access.

🚀 Technical Settings

  • Port: 3030

  • Endpoint: /mcp

  • Method: POST

  • Authentication: Authorization: Bearer senha-facil-123

  • Protocol: JSON-RPC 2.0 (Adapted for HTTP)

Related MCP server: local-env-mcp

🧪 Testing Remotely with CURL

You can validate the server's operation from any machine with network access (or via Wireguard VPN) using the commands below.

1. List Available Tools

curl -X POST http://10.7.0.1:3030/mcp \
  -H "Authorization: Bearer senha-facil-123" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/list","id":1}'

2. Call the 'hello' Tool

curl -X POST http://10.7.0.1:3030/mcp \
  -H "Authorization: Bearer senha-facil-123" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"hello","arguments":{"name":"Fabão"}},"id":2}'

♊ Configuring Gemini-CLI (Remote)

For the gemini-cli on your local machine or another VPS to access this server, you must configure it in the Gemini CLI configuration file (usually at ~/.gemini/config.yaml or via command).

Since this server uses direct HTTP (and not the stdio standard), you can use a "wrapper" or configure access via SSE if the server is extended, or use the fetch plugin if available.

Configuration example (Assuming use of a stdio-to-http proxy):

# No gemini-cli


gemini mcp add remote-server --command "npx @modelcontextprotocol/inspector http://10.7.0.1:3030/mcp"

In the ~/.gemini/settings.json file

"mcpServers": {                                                                                                                      
    "local-mcp": {                                                                                                                                     
      "url": "http://192.168.1.100:3030/mcp",                                                                                                          
      "headers": {                                                                                                                                     
        "Authorization": "Bearer senha-facil-123"                                                                                                      
      }                                                                                                                                                
    }             
  }

Note: Replace the IP with the correct address (e.g., 10.7.0.1 from the VPN).

🤖 Configuring Claude Code

Claude Code (and Claude Desktop) usually expect a local command. To access this remote server:

  1. Edit the claude_desktop_config.json file:

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

    • Linux: ~/.config/Claude/claude_desktop_config.json

  2. Add the configuration:

{
  "mcpServers": {
    "mcp-simple-server": {
      "command": "curl",
      "args": [
        "-s",
        "-X", "POST",
        "http://10.7.0.1:3030/mcp",
        "-H", "Authorization: Bearer senha-facil-123",
        "-H", "Content-Type: application/json",
        "-d", "{\"jsonrpc\":\"2.0\",\"method\":\"tools/call\",\"params\":<params>}"
      ]
    }
  }
}

Tip: For a more robust integration, it is recommended to use mcp-proxy which converts stdio to your server's HTTP API.

💬 Configuring ChatGPT (GPT Codex / Actions)

To use this server as an Action in ChatGPT:

  1. Go to Custom GPTs -> Create a GPT -> Configure -> Create new action.

  2. In the Authentication field, choose API Key, select Bearer, and enter: senha-facil-123.

  3. In the Schema field, use an OpenAPI specification (simplified example below):

openapi: 3.0.0
info:
  title: MCP Simple Server
  version: 1.0.0
servers:
  - url: http://195.35.42.89:3030 # IP Público ou DNS
paths:
  /mcp:
    post:
      operationId: callMcpTool
      summary: Chama uma ferramenta do MCP
      requestBody:
        content:
          application/json:
            schema:
              type: object
              properties:
                method: {type: string, example: "tools/call"}
                params: {type: object}
                id: {type: integer, example: 1}
      responses:
        '200':
          description: Sucesso

🐳 Docker

This project is ready to be run in containers.

1. Build the Image

To generate the Docker image locally:

docker build -t mcp-simple-server .

2. Start the Container (Docker Compose)

The recommended way to run it is by using the included docker-compose.yml, which configures the network and environment variables:

docker compose up -d

The container will be created with the name mcp-simple-server and will be available on port 3030.

3. Check Status and Logs

# Ver se o container está rodando
docker ps | grep mcp-simple-server

# Ver logs do servidor
docker compose logs -f

🛠️ Development

To run the server locally:

cd ~/apps/mcp-simple-server
npm install
node server.js

The server will be listening at http://localhost:3030.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivitySlowing
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • mempalace-mcp-http

    flejz
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables remote access to the MemPalace MCP server via HTTP, supporting bearer token authentication and concurrent clients while exposing all mempalace tools.
    -

  • Levitate

    iomz
    A
    license
    Not graded
    quality
    A
    maintenance
    Lifts local stdio MCP servers into remote Streamable HTTP endpoints for cloud-hosted AI clients, with bearer-token auth and tool policy filtering.
    4 npm
    MIT

  • mcp2api

    megamen32
    F
    license
    Not graded
    quality
    C
    maintenance
    Converts any MCP server into simple HTTP endpoints, enabling AI agents that only support HTTP (like ChatGPT Custom GPT or web_fetch) to use MCP tools.
    -