The developer platform for Model Context Protocol

MCP Kit simplifies the creation, management, and deployment of MCP servers for developers.

OpenAI

OpenAI

Anthropic

Anthropic

Google

Google

MCP Servers2 Active
4 Tools
8 Resources

Server Management

Create and configure MCP servers with intuitive controls.

Resources12 Files

Resource Management

Upload and organize files, databases, and web content.

Prompts5 Templates
You are a helpful assistant...
+ Variables: $context, $query

Prompt Templates

Create reusable prompt templates with variables.

Tools3 Connected

Tool Building

Connect APIs to create tools that extend LLM capabilities.

CONNECT YOUR APPS TO LLMS WITH EASE

What is MCP?

Model Context Protocol: The bridge between your applications and LLMs

If you've used Cursor in agent mode, you already understand the power of Model Context Protocol. It enables application developers to easily connect their programs or APIs to LLMs.

{
"protocol": "mcp",
"version": "1.0",
"methods": ["chat", "complete", "embed"],
"transport": "streamable-http"
}

Standardized protocol

A unified specification for LLM interactions that works across different models and providers.

Your App
MCP
LLM
Bidirectional communication

Seamless integration

Connect your existing applications to any LLM with minimal code changes.

Tool Calling
API Access
Context
Knowledge Base
Memory
Conversation History
Planning
Multi-step Tasks

Agentic capabilities

Transform LLMs into powerful agents that can use tools, access data, and execute complex tasks.

What is MCP Kit?

MCP Kit is a platform that simplifies the creation, management, and deployment of MCP servers. Configure your servers, resources, prompts, and tools through an intuitive web UI.

Build once, use everywhere

After building your MCP servers in our UI, integrate them into your applications with just a few lines of code:

// Connect to your MCP server
import { MCPClient } from '@mcpkit/client';

const client = new MCPClient({
serverUrl: 'https://your-mcp-server.mcpkit.com',
apiKey: process.env.MCP_API_KEY
});

// Use in your application
const response = await client.chat({
messages: [{
role: 'user',
content: userQuery
}],
tools: ['search', 'calculator']
});

Server Management

Configure and deploy multiple MCP servers with just a few clicks. Monitor performance, manage resources, and scale as needed.

MCP Servers

Production
Staging
Development

Production

Requests
1243
Users
87
Tools
6
+1
Resources
12
+7
Prompts
8
+3
Last updated: 2 minutes ago

Resource Management

Upload, organize, and manage resources that provide context to your LLMs. Support for documents, code, and structured data.

Resources

product-spec.pdf
2.4 MB2 days ago
api-schema.json
34 KB5 hours ago
user-flow.png
1.2 MB1 day ago
training-data.csv
5.7 MB3 days ago
model-config.yaml
12 KBJust now
5 resources • 9.3 MB used

Prompt Templates

Create reusable prompt templates with variables that can be used across your applications. Define consistent interactions for your LLMs.

Prompt Templates

Use {{variable_name}} syntax for variables
2 defined
topic
The main subject area
query
User's specific question
Preview
You are a helpful assistant that provides information about <span class="bg-primary/20 text-primary-foreground px-1 rounded">topic</span>. The user is asking: <span class="bg-primary/20 text-primary-foreground px-1 rounded">query</span>

Tool Development

Visually connect APIs to create tools that extend your LLM's capabilities. Map tool parameters to API variables without writing code.

Tool Builder

API Endpoint
Connected
GET
https://api.weather.example/v1/current
Parameter
location
Type
string
Required
Yes
Tool Definition
Name
getWeather
Description
Get current weather for a location

MCP is Gaining Momentum

The Model Context Protocol ecosystem is growing rapidly. Keep up with MCP Kit.

Sam Altman

Sam Altman

@sama

people love MCP and we are excited to add support across our products.

available today in the agents SDK and support for chatgpt desktop app + responses api coming soon!

2:02 PM · Mar 26, 2025 · 1.3M Views

Alex Albert

Alex Albert

@alexalbert__

A new version of the MCP spec was finalized today.

Some of the major changes:
• Auth framework based on OAuth 2.1
• Replaced the previous HTTP+SSE transport with Streamable HTTP transport
• Support for JSON-RPC batching
• Tool annotations for better describing tool behavior

10:49 AM · Mar 26, 2025 · 138.4K Views

Ready to build with MCP Kit?

Join thousands of developers building the next generation of AI-powered applications.

Get in Touch

Have questions about MCP Kit? Our team is here to help.

Contact Information

Email

hello@mcpkit.com

Discord

Join our community

GitHub

Star us on GitHub

Send us a message