AI & Machine Learning Bearer Token

MCP Servers REST API

Model Context Protocol servers for AI agents & assistants

MCP Servers provide standardized interfaces for AI agents and assistants to interact with external tools, data sources, and services through the Model Context Protocol. Developers use this API to create, manage, and orchestrate MCP-compliant servers that enable AI models to access file systems, databases, APIs, and custom business logic in a secure, controlled manner.

Base URL https://api.mcps.rest/v1

API Endpoints

Method	Endpoint	Description
GET	`/servers`	List all registered MCP servers with their capabilities and status
POST	`/servers`	Register a new MCP server with specified tools and resources
GET	`/servers/{serverId}`	Get detailed information about a specific MCP server
PUT	`/servers/{serverId}`	Update MCP server configuration, tools, or resources
DELETE	`/servers/{serverId}`	Unregister and remove an MCP server
GET	`/servers/{serverId}/tools`	List all tools exposed by a specific MCP server
POST	`/servers/{serverId}/tools/{toolName}/invoke`	Invoke a specific tool on an MCP server with parameters
GET	`/servers/{serverId}/resources`	List all resources accessible through an MCP server
GET	`/servers/{serverId}/resources/{resourceUri}`	Read content from a specific resource via MCP server
GET	`/servers/{serverId}/prompts`	List all prompt templates available on an MCP server
POST	`/servers/{serverId}/prompts/{promptName}`	Generate a prompt from a template with arguments
POST	`/sessions`	Create a new MCP session for agent interactions
GET	`/sessions/{sessionId}`	Get session details and connection status
DELETE	`/sessions/{sessionId}`	Close an active MCP session and cleanup resources
GET	`/health`	Check API health and server availability status

Code Examples

curl -X POST https://api.mcps.rest/v1/servers/fs-server/tools/read_file/invoke \
  -H 'Authorization: Bearer mcp_sk_1234567890abcdef' \
  -H 'Content-Type: application/json' \
  -d '{
    "arguments": {
      "path": "/data/config.json"
    }
  }'

const response = await fetch('https://api.mcps.rest/v1/servers/fs-server/tools/read_file/invoke', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer mcp_sk_1234567890abcdef',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    arguments: {
      path: '/data/config.json'
    }
  })
});

const result = await response.json();
console.log(result);

import requests

url = 'https://api.mcps.rest/v1/servers/fs-server/tools/read_file/invoke'
headers = {
    'Authorization': 'Bearer mcp_sk_1234567890abcdef',
    'Content-Type': 'application/json'
}
payload = {
    'arguments': {
        'path': '/data/config.json'
    }
}

response = requests.post(url, json=payload, headers=headers)
print(response.json())

Use MCP Servers from Claude / Cursor / ChatGPT

Get a hosted MCP endpoint for MCP Servers. Paste your MCP Servers API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls MCP Servers directly with your credentials — no local install, works on mobile.

list_mcp_servers Retrieve all available MCP servers with their capabilities and active tools

invoke_mcp_tool Execute a specific tool on an MCP server with provided arguments and return results

read_mcp_resource Access and read content from resources exposed by MCP servers (files, databases, APIs)

create_mcp_session Initialize a new MCP session for persistent agent-server interactions

generate_mcp_prompt Generate contextualized prompts from MCP server templates with dynamic arguments

Connect in 60 seconds

Paste your MCP Servers key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.

Connect MCP Servers to your AI →

MCP Servers REST API

API Endpoints

Sponsor this page

Code Examples

Use MCP Servers from Claude / Cursor / ChatGPT

Connect in 60 seconds

Related APIs