Deploy, manage, and access your MCP servers through a unified, secure platform.
Linkpoint simplifies deploying and managing your AI infrastructure:
Launch MCP servers with one click from our directory or use your custom container images on managed infrastructure.
Access all deployed servers via a single gateway URL and manage deployments, keys, and status from one dashboard.
Connect your end-user applications, chat clients, and CLI tools to the gateway to access your deployed AI capabilities securely.
Linkpoint provides a comprehensive set of features for your AI infrastructure:
Deploy servers instantly from the directory or configure custom deployments with ease.
Monitor status, manage configurations, and control deployed machines from the dashboard.
Single endpoint access to all your deployed servers, simplifying client configuration.
Generate API keys with granular permissions for specific servers and operations.
Servers run on scalable cloud infrastructure (Fly.io) managed via the Linkpoint platform.
Gain insights into request volume, performance, and usage patterns across your servers.
Understanding the workflow for deploying and accessing servers via Linkpoint:
Access the Linkpoint platform dashboard.
Navigate to 'Servers', choose from the Directory or 'Add Server' for custom deployment. Configure and launch.
Go to 'API Keys', create a new key, and assign permissions to your newly deployed server(s). Copy the key.
Use the provided Linkpoint Gateway URL and your generated API Key in your client application (e.g., Cursor, VS Code).
Your client connects to the Linkpoint Gateway, which authenticates and routes requests to the correct deployed server.
Use the Linkpoint dashboard to monitor server status, manage deployments, and update API keys.
Follow these steps to deploy your first server and connect your client:
Create an account or log in to the Linkpoint dashboard.
Go to the 'Servers' section. Click 'Add Server'. Choose a server from the Directory (e.g., 'MCP Memory Server') or configure a custom deployment using your own Docker image. Follow the prompts to configure resources and deploy.
Navigate to 'API Keys' and click 'Create API Key'. Give it a name and assign access to the server(s) you just deployed. Copy the generated API key - you'll need it for your client.
Open your MCP client (e.g., Cursor Settings -> Model Context Protocol). Enter the Linkpoint Gateway URL (found on your dashboard) and paste the API Key you created. Ensure the 'Authorization Header' is set (usually 'Authorization' by default).
Your client should now be connected through the Linkpoint Gateway to your deployed server. You can manage your deployments and keys anytime from the dashboard.
Configure popular clients to use the Linkpoint Gateway:
Configure your AI chat applications to connect to MCP servers:
{
"mcpServers": {
"my-server": {
"url": "https://your-gateway-url.com/sse",
"headers": {
"Authorization": "Bearer your-api-key-here"
}
}
}
}Use the MCP SDK to build applications that connect to MCP servers:
from mcp import ClientSession
from mcp.client.sse import sse_client
import asyncio
async def main():
# Create SSE client with URL and headers
async with sse_client(
url="https://your-gateway-url.com/sse",
headers={"Authorization": "Bearer your-api-key-here"}
) as (read, write):
# Create client session with the read/write streams
async with ClientSession(read, write) as session:
# Initialize the connection
await session.initialize()
# Now you can use the session to interact with the server
# For example, list available tools
tools = await session.list_tools()
print(f"Available tools: {tools}")
# The connection will be automatically closed when exiting the context managers
if __name__ == "__main__":
asyncio.run(main())