MCP server for Yandex Cloud

MCP server for Yandex Cloud is an application that provides AI agents
with access to manage Yandex Cloud resources via the standard
Model Context Protocol (MCP).
The server is deployed on a virtual machine in Yandex Cloud.

During installation, you select one of 11 available MCP servers:

• toolkit — manage Compute, VPC, IAM, Object Storage, YDB, and other services (42 tools).
• docs — search and navigate Yandex Cloud documentation.
• search — full-text search across resources via Yandex Search API.
• functions — manage Cloud Functions (serverless).
• containers — manage Serverless Containers.
• triggers — manage triggers for Cloud Functions and Serverless Containers.
• workflows — manage Yandex Cloud Workflows.
• apigateway — manage API Gateway.
• mcpgateway — manage MCP Gateway.
• datacatalog-consumer — access Data Catalog.

Key Features and Benefits

• Yandex AI Studio Integration. The MCP server integrates with AI Studio MCP Servers — connect it to your AI agents in just a few clicks.
• 11 MCP Servers. A wide range of servers for managing various Yandex Cloud services — from compute resources to serverless functions.
• Service Account Authentication. No additional API keys required — the application uses the VM’s service account for Yandex Cloud authentication.
• Bearer Token Authorization. All requests to the MCP server are protected with a Bearer token stored in Yandex Lockbox.
• Standard MCP Protocol. Uses the open Model Context Protocol for AI model interaction, ensuring compatibility with a wide range of clients.
• HTTP Transport. The MCP server is accessible on port 8000 (HTTP), simplifying integration with various clients and tools.
• Minimal Resources. The application runs on a virtual machine with minimal configuration (2 vCPU, 2 GB RAM), reducing infrastructure costs.
• Docker Containerization. The server runs in Docker containers (MCP + nginx auth proxy).

1. Create a secret in Yandex Lockbox to store the Bearer token for authorization. The secret must contain a bearer_token key with an arbitrary token value.
2. Create a service account with the lockbox.payloadViewer role and the roles required for the selected MCP server:
   • toolkit — editor or compute.admin
   • docs — no additional roles required
   • search — search-api.webSearch.user
   • functions — editor or functions.admin
   • containers — editor or serverless-containers.admin
   • triggers — editor or functions.admin
   • workflows — editor or serverless.workflows.admin
   • apigateway — editor or api-gateway.admin
   • mcpgateway — editor or serverless.mcpGateways.admin
   • datacatalog-consumer — data-catalog.viewer
3. In the management console select the Cloud Apps service.
4. In the left panel, select Application Store.
5. Select MCP Yandex Cloud and click the Use button.
6. Specify:
   • Application name.
   • (Optional) Application description.
   • The service account created earlier.
   • The cloud subnet where the virtual machine will be deployed.
   • MCP Server — select the desired server from the list (toolkit, docs, search, etc.).
   • (Optional) Folder ID — Yandex Cloud folder ID. Used as the default value for tools. If omitted, the AI agent will ask for it when needed.
   • Lockbox secret ID containing the Bearer token for authorization.
7. Click the Install button and wait for the application to install.
8. After installation, the MCP server will be available at http://<vm-ip>:8000.
9. Configure your AI client (Claude Desktop, Cursor, etc.) to connect to the MCP server, specifying the Authorization: Bearer <your-token> header.

Connecting the MCP server to Yandex AI Studio

1. In the management console open the AI Studio service and go to the MCP Hub section.
2. Click Create MCP server and choose the External MCP server type.
3. Specify the server name and the ID of the VPC network where the MCP server virtual machine is deployed. The network ID is available on the application page in the Application data section, VPC network ID field.
4. At the MCP server connection step, fill in the parameters with the values from the Application data section:
   • MCP server type — value from the MCP server type field.
   • MCP server URL — value from the MCP server URL (internal) field.
   • In the HTTP headers block, add an Authorization header with the value Bearer <bearer_token>, where <bearer_token> is the token from the Yandex Lockbox secret specified during application installation.
5. Click Connect. After a successful connection check, select the required tools and click Connect — the MCP server will be registered in AI Studio and available for use by AI agents.

• Managing virtual machines, networks, and disks via an AI assistant (toolkit).
• Searching Yandex Cloud documentation for up-to-date service information (docs).
• Creating and managing serverless functions via natural language (functions).
• Managing triggers, API Gateway, and Serverless Containers (triggers, apigateway, containers).
• Searching cloud resources via Yandex Search API (search).
• Managing MCP Gateway and Workflows (mcpgateway, workflows).

Yandex Cloud technical support is available 24/7. The types of requests you can submit and the appropriate response time depend on your pricing plan. You can switch to the paid support plan in the management console. You can learn more about the technical support terms here.