# Yandex Tracker [Health: Active]

**Category:** 📋 Product Management  
**Repository:** https://github.com/sdamarketing/tracker_mcp  
**GitHub Stars:** 1  
**npm Downloads (last month):** 1038  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/yandex-tracker

## Description
MCP server for Yandex Tracker: full API v3 coverage (187 tools)

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "yandex-tracker": {
    "command": "npx",
    "args": ["-y","tracker-mcp"],
    "env": {
      "TRACKER_TOKEN": "",
      "TRACKER_ORG_ID": ""
    }
  }
}
```

**Requires environment variables:** `TRACKER_TOKEN`, `TRACKER_ORG_ID` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Yandex Tracker MCP server does

The Yandex Tracker MCP server gives an AI assistant access to Yandex Tracker through the Model Context Protocol. The implementation lists 187 tools covering the Tracker API v3, including issue search and editing, status changes, links, comments, checklists, worklogs, attachments, queues, boards, sprints, dashboards, projects, goals, portfolios, automation, reports, filters, administration, and migration operations.

The available operations include both retrieval and mutation. Agents can search with Tracker filters and query language, create or update issues, manage comments and attachments, perform bulk changes, and work with administrative data such as absences or historical imports. The exact actions remain subject to the permissions of the configured Tracker credentials.

## How it works

An MCP client starts the server over stdio, or the packaged server can be run in HTTP mode. The server authenticates against Yandex Tracker using a Tracker token and organization ID, then translates MCP tool calls into Tracker API requests.

In addition to tools, the Yandex Tracker MCP server provides resources addressed with `tracker://` URIs. These include issue context with recent comments, queue creation fields, board information, catalog values, and the current user profile. It also exposes prompts for workflows such as standups, weekly reports, triage, sprint reviews, issue closure, and issue creation.

Tools carry MCP annotations for read-only and destructive behavior. Delete and bulk tools require `confirm: true`; read-only clients can use the `TRACKER_READ_ONLY=1` setting to expose only the 80 read tools and hide write operations.

## Setup and configuration

The project requires Node.js 20 or newer for a manual installation. It can be installed globally from npm with `npm install -g tracker-mcp`, followed by `tracker-mcp setup`, or built from the repository with `npm install`, `npm run build`, and `npm run setup`. The setup wizard accepts the Tracker credentials, validates them against the API, and can configure selected AI clients.

A one-command installer is provided for macOS, Linux, and WSL. Docker images are also published for both stdio and HTTP usage. In stdio mode, the client launches the container; in HTTP mode, the container listens on port 3407 and can use `MCP_AUTH_TOKEN` for authentication.

Required configuration values are `TRACKER_TOKEN` and `TRACKER_ORG_ID`. The repository also includes a `yandex-tracker` agent skill that can be installed with `npx skills add sdamarketing/tracker_mcp`; it contains procedural guidance for Tracker queries, field values, and common API details.

## Tools and capabilities

The Yandex Tracker MCP server supports these areas:

- Search, create, edit, move, link, transition, and close issues.
- Manage comments, reactions, checklists, worklogs, files, and temporary files.
- Perform bulk issue changes and bulk project or goal updates.
- Create and maintain queues, boards, columns, sprints, dashboards, reports, and saved filters.
- Manage projects, goals, portfolios, permissions, events, metrics, and attachments.
- Configure automation triggers, actions, workflow statuses, and transitions.
- Read and write Tracker catalogs such as types, statuses, priorities, resolutions, fields, and categories.
- Import historical issues, comments, links, worklogs, and attachments through administrative tools.

## Limitations and notes

The server requires valid Yandex Tracker credentials and an organization ID. Operations that modify or remove data may require confirmation, and access is limited by the Tracker account’s permissions. IAM tokens used with Tracker have a stated lifetime of no more than 12 hours in the accompanying agent guidance.

Windows users are directed to use WSL with the installer or follow the manual setup process. Client support depends on whether the client can run MCP servers and, for prompts, whether it implements MCP prompt features.

_Full upstream README: https://allmcps.com/mcp/yandex-tracker/readme_