Skip to main content
Glama
yandex-direct-mcp-plus

yandex-direct-mcp-plus

by Pavelsiba

Изменить ключевые слова

update_keywords

Edit keyword phrase text, {param1}/{param2} variables, and autotargeting categories or brand options. Recheck keywords after changes because edits may create new IDs or remove duplicates.

Instructions

Изменить текст ключевых фраз, подстановочные переменные {param1}/{param2} и настройки автотаргетинга. Правка текста может привести к появлению фразы с новым ID или к её удалению как дубликата — сверьтесь с list_keywords после вызова. Автотаргетинг — фраза ---autotargeting: её ID берётся из list_keywords, категории запросов и брендов задаются autotargeting_categories / autotargeting_brand_options. Выключить автотаргетинг целиком — manage_keywords с suspend (delete Директ отбивает). Ставки меняет set_keyword_bids, статус — manage_keywords.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "keywords": "<keywords>"
    },
    "name": "update_keywords"
  }
}

Input Schema

NameRequiredDescriptionDefault
keywordsYesФразы и их новые значения; поля, которые не переданы, остаются прежними

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changedv2.1.0
    • changedInput schema / $schema
      Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • addedInput schema / additionalProperties
      Added value: +false
    • removedInput schema / properties / dry_run
      Removed value: -{
      -  "description": "true — ничего не менять: проверить параметры и вернуть тела запросов, которые ушли бы в Директ. Чтение при этом выполняется.",
      -  "type": "boolean"
      -}
    • addedInput schema / properties / keywords / items / properties / autotargeting_brand_options
      Added value: +{
      +  "description": "Запросы с брендами и без. Меняются только переданные флаги, остальные остаются прежними. Только для фразы ---autotargeting",
      +  "properties": {
      +    "with_advertiser_brand": {
      +      "description": "Запросы с брендом рекламодателя",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "with_competitors_brand": {
      +      "description": "Запросы с брендами конкурентов",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "without_brands": {
      +      "description": "Запросы без упоминания брендов",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    }
      +  },
      +  "type": "object"
      +}
    • addedInput schema / properties / keywords / items / properties / autotargeting_categories
      Added value: +{
      +  "description": "Категории запросов автотаргетинга. Меняются только переданные, остальные остаются прежними. Только для фразы ---autotargeting; текущие значения — list_keywords с autotargeting: true",
      +  "properties": {
      +    "accessory": {
      +      "description": "Сопутствующие запросы — товары, которые берут вместе с этим",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "alternative": {
      +      "description": "Альтернативные запросы — замена товара",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "broader": {
      +      "description": "Широкие запросы — общая потребность, под которую подходит товар",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "exact": {
      +      "description": "Целевые запросы — прямо описывают товар из объявления",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    },
      +    "narrow": {
      +      "description": "Узкие запросы — товар с уточнениями",
      +      "enum": [
      +        "YES",
      +        "NO"
      +      ],
      +      "type": "string"
      +    }
      +  },
      +  "type": "object"
      +}
    • changedInput schema / properties / keywords / items / properties / keyword_id / description
      Previous value: -"ID изменяемой ключевой фразы"New value: +"ID изменяемой ключевой фразы или автотаргетинга"
  2. Changed1 schema field changedv1.7.0
    • addedInput schema / properties / dry_run
      Added value: +{
      +  "description": "true — ничего не менять: проверить параметры и вернуть тела запросов, которые ушли бы в Директ. Чтение при этом выполняется.",
      +  "type": "boolean"
      +}
  3. First observedv1.6.1

TDQS

A4.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnlyHint=false and openWorldHint=true, so the description carries the real burden and does so well: editing text can yield a phrase with a new ID or silently delete it as a duplicate, autotargeting text is immutable, and only passed fields change. It does not address partial-failure semantics across a batch of up to 1000 phrases, which is the remaining behavioral gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with what the tool changes, then the ID-mutation hazard, then the autotargeting workflow, then sibling routing — a logical ordering with no filler. It is dense and long-ish, but every clause carries operational information; only the trailing bids/status sentence could be compressed.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a mutation tool with no output schema and a single deeply nested array parameter, the description covers the risky behaviors, the verification loop, and the alternative tools. Missing only return/partial-success expectations for a multi-item batch and any permission or rate considerations.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description earns above baseline by explaining the cross-field dependency an agent would otherwise miss — autotargeting_* fields apply only to the ---autotargeting phrase whose ID must first be obtained from list_keywords. It adds little on user_param1/user_param2 beyond the schema's own 'null clears' note.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource plus the exact editable surfaces: phrase text, {param1}/{param2} substitution variables, and autotargeting settings. It also names the sibling tools that handle adjacent concerns (set_keyword_bids for bids, manage_keywords for status/suspend), so an agent can distinguish it without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly routes adjacent operations elsewhere: bids go to set_keyword_bids, status changes and full autotargeting disablement go to manage_keywords (noting that Direct rejects delete for the autotargeting phrase). It also prescribes a post-call verification step via list_keywords, giving a complete when/when-not picture.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.