Searches Yandex through the Yandex Cloud Search API v2 and returns smart snippets, page excerpts long enough to answer from, with sync and async modes and a result cache.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .claude/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
skills CLI
$ npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .agents/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .cursor/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .gemini/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .github/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "yandex-search-api" agent skill from https://github.com/artwist-polyakov/polyakov-claude-skills/tree/main/plugins/yandex-search-api/skills/yandex-search-api into .opencode/skills/yandex-search-api/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "yandex-search-api", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
yandex-search-api
GitHub stars
208
Token cost
~3k tokens
SKILL.md length
923 words
Files
28 (incl. scripts, references)
Skills in repo
21
Repo updated
First seen
Licence
MIT
At a glance
Searches Yandex through the Yandex Cloud Search API v2 and returns smart snippets, page excerpts long enough to answer from, with sync and async modes and a result cache.
Works in 5 steps: Определи регион → Режим поиска берётся из config.json →… → Verify config: убедись, что заданы… → …
Researching a question using what Russian-language sites say about it
SKILL.md covers Инфоконтексты (Smart Snippets), Config, Workflow and Scripts, plus 5 more sections
Runs Shell scripts from its folder; calls bash and sh; reaches leroymerlin.ru; needs YANDEX_AI_API_KEY
What it does
This skill parses Yandex search results through the Yandex Cloud Search API v2, in synchronous and asynchronous modes. Its main feature is smart snippets: for each hit the API returns a passage of about 500 tokens relevant to the query, with quotes from the page, so one request can answer a question without opening each site. Smart snippets are on by default in sync mode. Texts are saved to a cache folder and only a one-line-per-document index is printed, so you read the saved file whole or grep it instead of repeating the search.
Smart snippets come with limits: sync mode only, the Russian search type only, at most 20 documents per request, and a higher price per request than a plain search. The `--no-snippets` flag or a config setting returns just positions and domains. Setup needs a Yandex Cloud service account with the `search-api.webSearch.user` role, using either an IAM authorized key or a service account API key. Before searching, the agent works out the region, looks up its ID with a bundled script, and asks you and waits if the region is unclear.
When your agent uses it
Researching a question using what Russian-language sites say about it
Getting Yandex result positions and domains for a query
Searching Yandex for a specific city or region
Reading saved search excerpts instead of re-running the same query
Example prompts
“Search Yandex for how VAT works under the simplified tax system in 2026 and summarize the excerpts.”
“Find Yandex results for this query in the Kazan region and read the saved pack.”
“Give me only positions and domains from Yandex for this phrase, without snippets.”
Requirements
A Yandex Cloud service account with the search-api.webSearch.user role
An IAM authorized key or a service account API key
Bash to run the bundled scripts
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
1Определи регион
2Режим поиска берётся из config.json → search.mode (по умолчанию sync).
3Verify config: убедись, что заданы yandex_cloud_folder_id,
4Run search с полученным region ID
5Present results: со snippets — прочитай пак cache/results/.md
What it can do on your machine
Read from SKILL.md and the folder at commit 8bbeead. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 8 files in scripts/ (Shell, from the files we listed), which the agent can run.
Shell commands in SKILL.md call:
bash
sh
From the folder's file list and the shell code blocks in SKILL.md.
Network
Hosts in commands or code, which the agent is likely to contact:
leroymerlin.ru
Also links to:
aistudio.yandex.ru
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names these keys or tokens, usually read from environment variables:
YANDEX_AI_API_KEY
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Yandex Search API loads about 3k tokens when it runs, and up to ~6.4k if it reads all its reference files. Until then it costs about 104 tokens; SKILL.md has 923 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~104
When it runs· the whole SKILL.md, loaded when a task matches
~3k
With references· SKILL.md plus every file in references/, read only if the agent opens them
~6.4k
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check: notes
The automated check noted patterns worth knowing about, such as sudo or a known installer.
NoteMentions a .env fileSKILL.md:59
ig.json` и `YANDEX_AI_API_KEY` в `config/.env`.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
Download SKILL.mdSave it as .claude/skills/yandex-search-api/SKILL.md (or your agent's skills folder). This skill also uses 27 other files; get the full folder from GitHub.
name
yandex-search-api
description
Поиск в Яндексе через Yandex Cloud Search API v2 с содержательными
выдержками найденных страниц (smart snippets) — материал для ответа,
а не только ссылки. Синхронный и асинхронный режимы, кэш результатов.
Triggers: yandex search api, поиск в яндексе, выдача яндекса,
serp яндекс, парсинг выдачи, smart snippets, смарт-сниппеты,
инфоконтексты, выдержки страниц, найди в яндексе, что пишут в рунете.
yandex-search-api
Parse Yandex SERP via Yandex Cloud Search API v2 (sync + async).
Инфоконтексты (Smart Snippets)
Search API отдаёт не только ссылку и сниппет в пару предложений, а фрагмент
найденной страницы примерно на 500 токенов — релевантный запросу и с цитатами
из документа. Информации уже достаточно, чтобы ответить, а обрабатывать
страницу самому не нужно.
Это то, ради чего скилл ставят в агента: одного запроса хватает на ответ, а не
только на список ссылок.
Включено по умолчанию, отдельного скрипта не нужно:
bash
bash scripts/web_search_sync.sh --query "как считается НДС для УСН в 2026" --region-id 225
Тексты в ответ не печатаются — они уходят в cache/results/<hash>.md, а в
stdout приходит индекс по строке на документ. Дальше читай пак файлом: целиком,
если нужен весь материал, или грепом, если нужен один источник. Повторный поиск
той же фразы ради текста, который уже лежит в паке, — потраченные деньги.
Три ограничения, которые меняют выбор команды:
Только синхронный режим.web_search_async.sh инфоконтекстов не вернёт.
Только SEARCH_TYPE_RU. С другим типом поиска скилл выполнит запрос, но
без инфоконтекстов, и скажет об этом.
Не больше 20 документов. Попросить больше нельзя: при --results выше 20
API отдаёт не 20, а вдвое меньше — по той же цене. Скилл клампит сам.
Дороже. 1 500 ₽ за 1000 запросов против 488 ₽ у обычного синхронного.
Когда нужны только позиции и домены — --no-snippets (или
search.smart_snippets.enabled: false в конфиге).
Для работы нужен сервисный аккаунт Яндекс.Облака с ролью search-api.webSearch.user.
Поддерживаются IAM через авторизованный JSON-ключ и API-ключ сервисного аккаунта.
Пошаговая настройка обоих вариантов: config/README.md.
Существующие IAM-конфигурации работают без изменений. Для API-ключа задайте
auth.mode: "api_key" в config.json и YANDEX_AI_API_KEY в config/.env.
Workflow
STOP! Before any search:
Определи регион:
Если пользователь указал город/регион — найди ID автоматически:bash
bash scripts/search_region.sh --name "Казань"
Если регион не понятен из контекста — СПРОСИ и ЖДИ ответа:
"Для какого региона искать?
- Вся Россия (по умолчанию)
- Москва
- Конкретный город (какой?)"
НЕ ПРОДОЛЖАЙ пока пользователь не ответит!
Полученное название → search_region.sh --name "..." → получаешь ID
Для неоднозначных случаев (Москва город vs область) — уточни у пользователя
Режим поиска берётся из config.json → search.mode (по умолчанию sync).
Не спрашивай — используй то, что в конфиге.
Исключение: если нужны инфоконтексты, режим всегда sync, а тип поиска —
SEARCH_TYPE_RU; в асинхронном API и в других поисковых базах их нет.
Verify config: убедись, что заданы yandex_cloud_folder_id,
соответствующий credential (поле auth.mode необязательно). Не печатай ключ или токен.
Run search с полученным region ID
Present results: со snippets — прочитай пак cache/results/<hash>.md
и отвечай по нему, ссылаясь на источники; без snippets — позиция, заголовок,
URL, сниппет
Scripts
iam_token_get.sh
Generate or validate IAM token from Service Account key. Needed only when
auth.mode is iam.
bash
bash scripts/iam_token_get.sh
Token is cached in cache/iam_token.json and auto-refreshed when expired.
web_search_sync.sh
Synchronous search — one query at a time, immediate results.
Единственный режим, который умеет smart snippets.
bash
# Single query — с выдержками страниц
bash scripts/web_search_sync.sh \
--query "купить дымоход" \
--region-id 213
# Только позиции и ссылки, без выдержек
bash scripts/web_search_sync.sh \
--query "купить дымоход" \
--region-id 213 \
--no-snippets
# Batch from file — по строке на запрос в stdout, пак на каждый запрос
bash scripts/web_search_sync.sh \
--file queries.txt \
--region-id 225
Results saved to cache/results/<hash>.md (пак с выдержками),
cache/results/<hash>.json (parsed) и cache/results/<hash>.raw — исходное
тело ответа: JSON с инфоконтекстами, XML без них.
Show full SKILL.md (379 more words)Show less
web_search_async.sh
Asynchronous batch search — submit many queries, poll for results.
bash
# Submit batch and wait
bash scripts/web_search_async.sh \
--file queries.txt \
--region-id 213
# Resume after timeout/interrupt
bash scripts/web_search_async.sh --resume
Param
Required
Default
Values
--file, -f
yes
-
File with queries
--region-id, -r
no
from config (225)
Region ID
--poll-interval
no
10
Poll interval (minutes)
--max-wait
no
120
Max wait before timeout (minutes)
--resume
no
-
Continue polling pending ops
Async workflow:
Script submits all queries as async operations
Polls every poll_interval minutes for completion
Downloads and parses results as they complete
If max_wait exceeded: prints summary + resume command
On restart with --resume: continues from cache/ops/, no duplicates
NOTE for agent: Async execution can take minutes to hours.
The script handles polling automatically. If it times out,
re-run with --resume to continue.
Инфоконтексты в асинхронном режиме недоступны — вернутся только ссылки и
короткие сниппеты. Нужны фрагменты страниц — web_search_sync.sh --file.
Зато async в ~16 раз дешевле синхронного и в ~50 раз дешевле инфоконтекстов.
regions_tree.sh
Show common region IDs.
bash
bash scripts/regions_tree.sh
search_region.sh
Find region ID by name.
bash
bash scripts/search_region.sh --name "Казань"
Output Format
Each search result contains:
position — rank in SERP
title — page title
url — page URL
snippet — text snippet (up to 300 chars)
domain — site domain
extract — инфоконтекст: фрагмент страницы; пустая строка, если его нет
Results cached in cache/results/:
<hash>.md — пак: заголовок, URL и инфоконтекст по каждому документу
<hash>.raw — исходное тело ответа: XML для обычной выдачи, JSON для
инфоконтекстов (формат выбирает сервер)
<hash>.json — parsed JSON array, единый формат для обоих случаев
<hash> — первые 12 символов md5 от текста запроса.
Не печатай инфоконтексты в stdout. 20 документов по ~500 токенов не
помещаются в буфер вывода песочницы, и скрипт упадёт молча. Скрипты печатают
индекс, тексты читаются из пака через Read/grep.
Tests
Офлайн-тесты: ни сети, ни сервисного аккаунта.
bash
sh scripts/tests/run.sh
Проверяют сборку тела запроса, разбор обоих форматов ответа (XML и JSON),
отрисовку пака и разбор флагов CLI. Тело запроса можно посмотреть и вручную,
ничего не оплачивая:
User: Разберись, как в 2026 считается НДС для УСН, и дай ответ со ссылками
Claude: [Регион из контекста не следует, но вопрос общероссийский → 225]
[Проверяет конфигурацию авторизации без вывода секрета]
[Ищет с выдержками — они включены по умолчанию]
bash scripts/web_search_sync.sh --query "НДС для УСН 2026 порядок расчёта" --region-id 225
=== Results for: НДС для УСН 2026 порядок расчёта ===
Region: 225 | Page: 0 | Snippets: true
# символов домен заголовок
1 1840 nalog.gov.ru Об изменениях в главе 26.2 НК РФ
2 1512 consultant.ru НДС при УСН с 2026 года
3 0 example-blog.ru Что изменится для упрощенцев
...
Всего: 20, с выдержками: 17
Пак: cache/results/a1b2c3d4e5f6.md (48 KB) — читай его вместо повторного поиска
[Читает пак и отвечает по нему, а не по заголовкам]
Read cache/results/a1b2c3d4e5f6.md
По порогу выручки: ... (nalog.gov.ru)
По ставкам: ... (consultant.ru)
User: Найди топ выдачи яндекса по запросу "купить сэндвич дымоход" в Москве
Claude: Уточню: Москва-город или Москва и область?
User: Москва город
Claude: [Находит ID региона]
bash scripts/search_region.sh --name "Москва"
→ Москва = 213
[Проверяет конфигурацию авторизации без вывода секрета]
[Задача про позиции в выдаче, содержание страниц не нужно —
отключает инфоконтексты: дешевле втрое]
bash scripts/web_search_sync.sh --query "купить сэндвич дымоход" \
--region-id 213 --no-snippets
=== Results for: купить сэндвич дымоход ===
Region: 213 | Page: 0 | Snippets: false
1. Сэндвич-дымоходы купить в Москве — Леруа Мерлен
https://leroymerlin.ru/...
Широкий ассортимент сэндвич-дымоходов...
2. Дымоходы сэндвич — купить в интернет-магазине
https://...
...
Всего: 10, с выдержками: 0
JSON: cache/results/7f3a91c2e5d8.json
Raw: cache/results/7f3a91c2e5d8.raw
User: Проверь выдачу по запросам из файла queries.txt в Казани
Claude: [Находит ID региона]
bash scripts/search_region.sh --name "Казань"
→ Казань = 43
[Проверяет конфигурацию авторизации без вывода секрета]
[Батч по одному запросу; в stdout — строка на запрос,
тексты и разбор лежат в файлах]
bash scripts/web_search_sync.sh --file queries.txt --region-id 43
запрос результаты пак
дымоход сэндвич цена 20 док., выдержек 18 cache/results/a1b2c3d4e5f6.md
купить дымоход казань 20 док., выдержек 15 cache/results/b2c3d4e5f6a1.md
дымоход нержавейка 20 док., выдержек 19 cache/results/c3d4e5f6a1b2.md
=== Batch complete: 3/3 queries processed ===
Yandex Search API next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Yandex Search API compared with similar skills
Skill
Stars
Used in
Tokens
Auto-check
Licence
Repo updated
Yandex Search API this skillartwist-polyakov/polyakov-claude-skills
Use the user's already signed-in official ChatGPT website account, especially GPT-5.5 Pro / ChatGPT Pro, to perform product research, market research, competitor research, second-opinion research…
Use the user's already signed-in official ChatGPT website account, especially GPT-5.5 Pro / ChatGPT Pro, to perform product research, market research, competitor research, second-opinion analysis…
Crawls a site with Crawl4AI to audit titles, meta tags, H1s, canonicals, navigation and internal links, and to compare landing pages and competitor sites.
Publishes already-built static pages to a GitHub Pages repo under a year, year-month and slug folder layout, optimizes large images and returns the public URL.
Shell scripts for web search and research through the Perplexity API: raw results, cited answers, background deep research and page fetching, with results cached on disk.
Publish static page artifacts to SourceCraft Sites (Yandex infrastructure, works in Russia), with advisory image optimization and an original-image path.
Searches Yandex through the Yandex Cloud Search API v2 and returns smart snippets, page excerpts long enough to answer from, with sync and async modes and a result cache. This skill parses Yandex search results through the Yandex Cloud Search API v2, in synchronous and asynchronous modes. Its main feature is smart snippets: for each hit the API returns a passage of about 500 tokens relevant to the query, with quotes from the page, so one request can answer a question without opening each site.
When should I use Yandex Search API?
Yandex Search API fits situations like: researching a question using what Russian-language sites say about it; getting Yandex result positions and domains for a query; searching Yandex for a specific city or region; reading saved search excerpts instead of re-running the same query.
How do I install Yandex Search API in Claude Code?
Run `npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a claude-code`. Or copy the skill folder (plugins/yandex-search-api/skills/yandex-search-api in artwist-polyakov/polyakov-claude-skills) into .claude/skills/yandex-search-api in your project. Claude Code loads it when a task matches its description.
How do I install Yandex Search API in Codex?
Run `npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a codex`. Or copy the skill folder (plugins/yandex-search-api/skills/yandex-search-api in artwist-polyakov/polyakov-claude-skills) into .agents/skills/yandex-search-api in your project. Codex loads it when a task matches its description.
Can I use Yandex Search API in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add artwist-polyakov/polyakov-claude-skills --skill yandex-search-api -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/yandex-search-api, .gemini/skills/yandex-search-api, .github/skills/yandex-search-api and .opencode/skills/yandex-search-api in your project.
What does Yandex Search API need to run?
Going by SKILL.md and its folder, Yandex Search API needs a shell for the scripts in its folder, the command-line tools its instructions call (bash and sh) and credentials named YANDEX_AI_API_KEY. Our summary lists: A Yandex Cloud service account with the search-api.webSearch.user role; An IAM authorized key or a service account API key; Bash to run the bundled scripts.
Does Yandex Search API access the network?
SKILL.md names 2 domains. In commands or code: leroymerlin.ru; the agent is likely to contact it when it follows the instructions. As links in the text: aistudio.yandex.ru. This is read from the text; nothing was executed.
Is Yandex Search API safe to install?
Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Yandex Search API use?
Yandex Search API is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Yandex Search API use?
About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 3.3k tokens, read only when the agent opens those files.
What are the alternatives to Yandex Search API?
Skills that share tags, products or a category with Yandex Search API: What Happened (kunchenguid/whathappened, 231 stars), Bright Data MCP (brightdata/skills, 264 stars), Chatgpt Web Research (bear2u/my-skills, 932 stars) and Gsc (jdrhyne/agent-skills, 240 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Yandex Search API?
artwist-polyakov (a GitHub user) maintains it in artwist-polyakov/polyakov-claude-skills, which has 208 GitHub stars. The repository holds 21 skills in this directory. The repository was last updated on October 8, 2026.
