Yandex Tracker

Yandex Tracker MCP Server with OAuth2 support

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "aikts-yandex-tracker-mcp": {
      "command": "uvx",
      "args": [
        "yandex-tracker-mcp@latest"
      ],
      "env": {
        "HOST": "0.0.0.0",
        "PORT": "8000",
        "REDIS_DB": "0",
        "TRANSPORT": "stdio",
        "REDIS_PORT": "6379",
        "OAUTH_STORE": "redis",
        "OAUTH_ENABLED": "true",
        "TRACKER_TOKEN": "your_tracker_token_here",
        "REDIS_ENDPOINT": "localhost",
        "REDIS_PASSWORD": "your_redis_password",
        "TRACKER_ORG_ID": "your_org_id_here",
        "OAUTH_CLIENT_ID": "your_oauth_client_id",
        "OAUTH_SERVER_URL": "https://oauth.yandex.ru",
        "OAUTH_TOKEN_TYPE": "Bearer",
        "OAUTH_USE_SCOPES": "true",
        "TRACKER_IAM_TOKEN": "your_iam_token_here",
        "TRACKER_SA_KEY_ID": "your_key_id",
        "OAUTH_CLIENT_SECRET": "your_oauth_client_secret",
        "REDIS_POOL_MAX_SIZE": "10",
        "TOOLS_CACHE_ENABLED": "true",
        "TRACKER_CLOUD_ORG_ID": "your_cloud_org_id_here",
        "MCP_SERVER_PUBLIC_URL": "https://your-server.com",
        "TOOLS_CACHE_REDIS_TTL": "3600",
        "TRACKER_SA_PRIVATE_KEY": "your_private_key",
        "TRACKER_SA_SERVICE_ACCOUNT_ID": "your_sa_id"
      }
    }
  }
}

This MCP server enables you to securely access Yandex Tracker data through AI-driven clients. It exposes a comprehensive set of operations for queues, issues, comments, worklogs, and searches, with flexible authentication and optional caching to improve performance.

How to use

You connect an MCP client to the Yandex Tracker MCP Server to perform real-time operations against Yandex Tracker data. Use the provided stdio or HTTP transports to integrate with your AI assistant. Start with one of the supported startup methods and then configure your client to authenticate using tokens or OAuth as described in the configuration section. After connecting, you can list queues, create and update issues, add comments, manage worklogs, search with advanced queries, and navigate workflows all through the MCP protocol.

How to install

Prerequisites: ensure the required tooling is available on your system. The MCP server can run via a local transport or over HTTP, and it supports OAuth and token-based authentication.

Install the MCP server usage toolings and prepare environment variables as shown in the configuration sections.
Start using a local stdio transport with UVX. The following commands illustrate common startup patterns.

uvx yandex-tracker-mcp@latest

Additional configuration and usage notes

Your MCP client should be configured with the appropriate authentication method. You can choose from static OAuth tokens, IAM tokens, or dynamic IAM/OAuth configurations. The server also supports an optional Redis-backed cache and an OAuth data store for token management.

Security and transport notes

You can limit access to specific queues using environment variables and fine-tune security policies. Transport options include stdio for local development and streamable-http for web-based integrations. When using OAuth, you may enable the provider and configure callback URLs so clients can obtain and refresh tokens.

Available tools

queues_get_all

List all queues with optional field filtering, pagination, and per_page control; honors queue limits.

queue_get_tags

Fetch all tags for a specific queue.

queue_get_versions

Retrieve available queue versions with details.

queue_get_fields

Get global and queue-local fields, including whether fields are required.

queue_get_metadata

Get detailed metadata for a queue, with optional expansion for related data.

users_get_all

List users in the organization with pagination and key details.

user_get

Fetch information about a specific user by login or UID.

user_get_current

Get information about the currently authenticated user.

users_search

Search users by login, email, or real name with fuzzy matching for names.

get_global_fields

Retrieve all global fields available across Tracker.

get_statuses

Return all available issue statuses.

get_issue_types

Return all available issue types.

get_priorities

Return all available issue priorities.

get_resolutions

Return all available issue resolutions.

issue_get

Retrieve a full issue by ID with optional description.

issue_get_url

Generate a web URL for an issue.

issue_get_comments

Fetch all comments for an issue.

issue_add_comment

Add a comment to an issue with optional mentions and followers.

issue_update_comment

Update an existing issue comment.

issue_delete_comment

Delete a comment from an issue.

issue_get_links

Get related issue links.

issue_get_worklogs

Retrieve worklog entries for issues.

issue_add_worklog

Add a worklog entry to an issue.

issue_update_worklog

Update a worklog entry for an issue.

issue_delete_worklog

Delete a worklog entry.

issue_get_attachments

List attachments for an issue.

issue_get_checklist

Get checklist items for an issue.

issue_get_transitions

List possible status transitions for an issue.

issue_execute_transition

Execute a status transition on an issue.

issue_close

Close an issue with a specified resolution.

issue_create

Create a new issue in a queue.

issue_update

Update fields on an existing issue.

issues_find

Search issues using Yandex Tracker Query Language.

issues_count

Count issues matching a Yandex Tracker query.

Built by

VeilStrat

AI signals for GTM teams