Yandex Metrika skill plugin wrapper
claude-skill-yandex-metrika
A Claude Code plugin bundling a Yandex Metrika skill: traffic, conversions, UTM, search engines and Direct costs, with caching
- 0 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The agent runs shell scripts that read the Metrika Reporting API with an OAuth token
- Stores responses in cache/ and exports CSV to disk
Install
- Works in
- Services
Manual install
mkdir -p ~/.claude/skills && cp -r skills/yandex-metrika ~/.claude/skills/Global install for all projects, from the README.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent runs shell scripts that read the Metrika Reporting API with an OAuth token
- Stores responses in cache/ and exports CSV to disk
This is third-party code. Review the repository files before installing.
What it does
The repository packages Aleksandr Poliakov's yandex-metrika skill into a standalone Claude Code plugin with a .claude-plugin/plugin.json manifest, so it can be installed with one command instead of digging through the original monorepo. The skill provides shell scripts for Reporting API reports: counters and goals list, traffic summary, conversions, UTM report, search engines, e-commerce, Direct clients and costs, period comparison. Every command caches responses under cache/ keyed by counter plus dates plus parameters and keeps terminal output short, with full data going to CSV or a file to avoid bloating the model's context.
Who it is for. For those who want to install a ready Metrika skill in Claude Code with one command, without cloning a large monorepo.
Good fit when
- You want a ready set of shell scripts for Metrika reports in Claude Code
- Caching matters, so the API is not hit repeatedly and the model's context stays clean
- It is easier to install one plugin than to find the right skill inside someone else's monorepo
Not a fit when
- You want the original and its updates directly: this is a separately saved copy with no guaranteed sync to upstream
- You need Direct or Webmaster support: only Metrika is covered here
Example request
Show the traffic summary and conversions for the metallik.ru counter over the last monthLimitations
This is the repository author's saved copy of Aleksandr Poliakov's skill (artwist-polyakov/polyakov-claude-skills), not the original: upstream updates are not pulled in automatically. A Yandex OAuth token with metrika:read scope is needed, placed in config/.env. The scripts require a bash environment.
How to disable. Delete the yandex-metrika folder from ~/.claude/skills or .claude/skills, or disable the plugin via claude --plugin-dir.
FAQ
Is this the official skill from Poliakov?
No, it is llirikkcoder's saved working copy under the same MIT license; the original lives at artwist-polyakov/polyakov-claude-skills.
What token scope is needed?
metrika:read; the walkthrough is in skills/yandex-metrika/config/README.md.
Discussion
No questions or reviews yet.
Related
All Yandex Metrica toolsA marketplace of 22 Claude Code plugins: Yandex Direct, Metrika, Webmaster, Wordstat, SEO crawling, publishing and research
A Claude Code skill for Yandex Metrica built on bash scripts over curl: preset reports, goals, counters and raw log exports
Yandex ecosystem plugin marketplace
claude-yandex-skills
Fifteen Claude Code plugins for Yandex services sharing one OAuth token: Metrica, Webmaster, Wordstat, Direct, Disk, Tracker and more
A skill and condensed Yandex Metrika API documentation for an agent: reports, Logs API, counter management, OAuth without leaking the token into chat
Five cache-first shell skills for Direct, Metrika, Webmaster, Wordstat and Search API, with Claude choosing which one to run
SEO audit plugin with Yandex-specific checks
claude-seo-pro
A Claude Code plugin with 15 skills and 10 parallel subagents: a full SEO audit including checks for Yandex Turbo Pages, Metrika and robots.txt
README in short
The README is upfront that the skill is taken from artwist-polyakov/polyakov-claude-skills and saved as a working copy. It shows three install paths (global, per-project, or as a plugin via a local marketplace), metrika:read token setup, and a short list of 11 shell scripts with caching and exact, unsampled data.
README badge
Original SKILL.md
--- name: yandex-metrika description: | Аналитика Yandex Metrika: трафик, конверсии, UTM, поисковые системы. Cache-first подход для гигиены контекстного окна. Triggers: яндекс метрика, yandex metrika, metrika analytics, метрика трафик, метрика конверсии, метрика отчет. --- # yandex-metrika Работа с Yandex Metrika Reporting API v1. Отчеты по трафику, конверсиям, UTM-меткам, поисковым системам. ## Config Требуется YANDEX_METRIKA_TOKEN в config/.env. Инструкция: config/README.md. ## Philosophy 1. Cache-first, конфигурационные данные (счетчики, цели, инфо) кешируются надолго. Отчеты кешируются по ключу counter+dates+params. Перед API-запросом всегда проверяем кеш. 2. Context window hygiene, stdout ограничен 30 строками. Полные данные в CSV/файл. 3. Точные данные, accuracy=1 (без сэмплирования), фильтр isRobot по умолчанию. 4. Атрибуция, дефолт lastsign (последний значимый источник). Спрашиваем пользователя при первом запуске.