SkillFoxx

yawiki: read-only Yandex Wiki MCP

yawiki

A Python read-only Yandex Wiki MCP server where each user signs in through a browser form with an IAM token

  • 0 ★ on the repository
  • Last commit
MCP server

Extended access

Why this level

  • An HTTP server reads Yandex Wiki pages, attachments and tables on the user's behalf
  • Takes Yandex Cloud IAM tokens through a login form, no write operations
All reasons and checks
Russian stack

zzlodeyy/yandex-wiki-mcpAuthor:ZzlodeyY

Install

Manual install

cp deploy/dev.env .env && uv sync && uv run yawiki

Local run, the server starts at http://localhost:8000.

Access and checks

How we review

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

  • An HTTP server reads Yandex Wiki pages, attachments and tables on the user's behalf
  • Takes Yandex Cloud IAM tokens through a login form, no write operations

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

What it does

The server provides four read-only tools: page content by URL or slug, a list of child pages for navigating the wiki structure, a list of files attached to a page, and a list of dynamic tables on a page. Its distinct feature is authorization: it runs as a remote HTTP MCP, and each user, not the server owner, signs in through a browser form entering their own Yandex Cloud IAM token and organization id, after which tools act on their behalf. It has a health check and a configurable wiki root path via an environment variable.

Who it is for. For teams that want a shared wiki-reading server for multiple users without sharing one common token

Good fit when

  • Several people need to read the wiki through one deployed server under their own access rights
  • You only need to view pages, attachments and tables without editing
  • You need a Docker deployment with a health check

Not a fit when

  • You need to create or edit pages: the server is read-only
  • You need a local stdio run without standing up an HTTP server

Example request

Show the content of the homepage/platforma-neuro page in Yandex Wiki and list its child pages

MCP

Transport
http
Authentication
OAuth

Environment variables

HOST
Listen address, defaults to 0.0.0.0
PORT
Server port, defaults to 8000
MCP_SERVER_PUBLIC_URL
The server's public URL for production
OAUTH_TOKEN_TYPE
Token type, defaults to Bearer
YANDEX_WIKI_BASE_SLUG
Wiki root path, e.g. homepage/platforma-neuro

Limitations

The project has no stars, a single author. It is read-only, with no write operations. Each user has to get their own IAM token via yc iam create-token or the Yandex Cloud console, and it has a limited lifetime.

How to disable. Stop it with docker compose down or stop the server process, and remove it from your MCP client config.

FAQ

Can the server edit pages?

No, it only reads content, child pages, attachments and tables.

Is there one shared token for all users?

No, each user signs in through their own form with their own IAM token; the server does not hold one shared key.

Discussion

No questions or reviews yet.

All Yandex Wiki tools
Russian stack

An MCP server with 28 tools for the Yandex Wiki API v1: pages, attachments, comments, tables and recovering deleted pages

MCP serverExtended access
Russian stack

An MCP server for Yandex Wiki: the same full-text search as the web UI, pages, comments, attachments and dynamic tables, 33 tools

MCP serverExtended access7
Russian stack

An MCP server for Yandex Wiki authenticated with a Yandex Cloud IAM token: 12 tools for pages, subpages, comments and attachments

MCP serverExtended access
Russian stack

A Yandex Wiki MCP server with built-in browser-based PKCE authorization, no manual OAuth app setup required

MCP serverExtended access
Russian stack

An MCP server for Yandex Wiki: read pages by slug or URL, create, update and append, with a read-only mode

MCP serverBroad access9

README in short

The README describes the server as a read-only Yandex Wiki MCP with browser-form IAM token authorization per user. It lists four tools, a local run via uv, a Docker run on port 8001 with a health check, and an environment variable table for host, port, public URL and wiki root path.

README badge

Report a problem