DataLens official MCP server

datalens-mcp

The official DataLens MCP server from the development team: turns the public DataLens API into three gateway tools instead of hundreds of separate MCP tools

  • 4 ★ on the repository
  • Last commit
  • MIT
MCP serverOfficialEditors’ pick

Extended access

Why this level

  • invoke_command can call any DataLens API command, including ones that change dashboards and datasets, within the IAM token's rights
  • The IAM or OAuth token grants organization-level DataLens access under the user's rights
All reasons and checks
Russian stack

datalens-tech/datalens-mcp

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add mcp/datalens-tech-datalens-mcp

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • invoke_command can call any DataLens API command, including ones that change dashboards and datasets, within the IAM token's rights
  • The IAM or OAuth token grants organization-level DataLens access under the user's rights

This is third-party code. Review the repository files before installing.

What it does

On startup the server fetches the public DataLens API's OpenAPI spec and turns every RPC endpoint into a callable command. Instead of registering hundreds of separate tools, which would flood the model's context, it exposes a gateway of three: list_commands lists every command with a one-line summary, describe_commands returns the full description and parameter schema for the commands you need, and invoke_command calls a command by name with parameters. Authorization against Yandex Cloud goes through the yc CLI: the server fetches an IAM token itself via yc iam create-token and refreshes it before it expires. For internal DataLens installations an OAuth token is supported instead of yc.

Who it is for. For DataLens analysts and developers who want agent access to the entire public API without parsing hundreds of separate methods.

Good fit when

  • You need agent access to DataLens dashboards, charts and datasets in Yandex Cloud
  • You run an internal DataLens installation and need OAuth access instead of yc
  • You want to avoid flooding the model's context with hundreds of separate MCP tools

Not a fit when

  • You have no yc CLI configured via yc init for the cloud setup, and no OAuth token for an internal one
  • You want a ready set of named tools per operation, not a gateway of three generic commands

Example request

Find DataLens commands for working with datasets and show how to call dashboard creation

MCP

Transport
stdio
Authentication
OAuth

Environment variables

DATALENS_ORG_ID required
Yandex Cloud organization id, required for the cloud installation
DATALENS_INSTALLATION
Installation type: cloud (default) or internal
DATALENS_OAUTH_TOKEN secret
OAuth token for an internal DataLens installation
DATALENS_YC_STATIC_AUTH
Use a static token from DATALENS_API_AUTH_HEADER instead of the yc CLI

Limitations

The yc-based IAM token lasts 12 hours; with a static token (DATALENS_YC_STATIC_AUTH) the user is responsible for refreshing it and restarting the server before it expires. API responses are size-limited (10 MiB for API calls, 20 MiB for OpenAPI), and data returned to the agent is marked untrusted_data and should not be treated as an instruction.

How to disable. Remove the datalens entry from mcpServers in your MCP client config.

FAQ

Do I need a separate API key?

No, for the cloud setup a configured yc CLI is enough; the server fetches the token itself.

Does it work with an internal DataLens installation?

Yes, via DATALENS_INSTALLATION=internal, DATALENS_API_URL and DATALENS_OAUTH_TOKEN instead of yc.

Discussion

No questions or reviews yet.

Russian stack

MCP server for the Yandex DataLens public API: 70 tools for dashboards, charts, datasets and connections

MCP serverExtended access8Repository stars
Russian stack

A Rust MCP server for the Yandex DataLens public API: 60 methods via a full catalog, typed wrappers for common operations and IAM auth

MCP serverExtended access5Repository stars

DataLens Dev MCP

datalens-dev-mcp

Russian stack

Local MCP server and Codex plugin for Yandex DataLens dashboard development: read, author, publish, back up and clean up objects

MCP serverBroad access7Repository stars
Russian stack

A translator from the cloud DataLens API to internal open-source DataLens 2.9.0 calls, plus a Streamable HTTP MCP server on top

MCP serverExtended access5Repository stars
Official

Yandex Cloud's official MCP server set: documentation, search, app deployment, serverless functions, containers and a data catalog

MCP serverBroad access49Repository stars
Official

Official Yandex demo templates: MCP gateways over cloud functions for YandexART, charts, PDF, arXiv, Kaggle and weather

MCP serverExtended access1Repository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialAuthenticationRepository starsMain difference
DataLens official MCP server (this card)yesOAuth4
Yandex DataLens public API MCP servernoOAuth560 public API methods
DataLens: a Java database MCP servernonot required0Postgres, MySQL, SQLite

README in short

The README explains the three-tool gateway scheme, yc CLI authorization and the static-token alternative, internal installation setup, three run modes (npx, global install, local build), a full environment variable table, and a detailed security-boundaries section: response size limits, rejecting unsafe redirects, and marking data as untrusted_data.

README badge

Report a problem