Действие с ключевыми фразами

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

Annotations already declare the tool is not read-only (readOnlyHint: false), destructive (destructiveHint: true), and not idempotent (idempotentHint: false). The description adds the specific actions (suspend, resume, delete), which aligns with these annotations and gives clarity on the operation types. However, it does not disclose additional behavioral traits such as reversibility, permission requirements, or side effects beyond what annotations and the schema enum already imply. The bar is lower due to annotations, so this is acceptable but not exceptional.

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

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

The description is a single, compact sentence that front-loads the core concept: performing an action on keywords by ID, followed by a list of the allowed actions. There is zero redundancy, and every part contributes to understanding. It is appropriately minimal for the tool's simplicity.

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

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

For a simple two-parameter tool with an enum and an array, the description, combined with the schema and annotations, provides enough for an agent to make a correct call. It mentions the action types and the ID nature, and the schema defines required fields. While there is no output schema, that is not required for a mutation operation. The description is adequate, though it could hint at the return value (e.g., a status) or the effect on multiple items, but these are minor gaps.

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

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

Schema description coverage is 50%: only the 'ids' parameter has a description, while 'action' does not. The description compensates by listing the three enum values for 'action' in its text, and mentions 'по id' to clarify that ids are the target keywords. This adds context beyond the raw schema, but it does not fully elaborate on the semantics of each action or any constraints on ids (e.g., whether the same id can be in multiple actions). Given the low-to-moderate coverage, the description adds some value but not comprehensive meaning.

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

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

The description clearly states the verb 'Выполняет действие' (performs an action) on 'ключевыми фразами' (key phrases) by ID, and explicitly lists the three possible actions: suspend, resume, or delete. This specifies the resource and operation precisely, though it does not explicitly contrast with sibling tools like ad_action or campaign_action. The resource name 'keyword' already differentiates it, so it's clear but not fully distinguishing.

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

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

The description provides no guidance on when to use this tool versus alternatives, no prerequisites, no context about required permissions, and no exclusions. It only states what the tool does, without any 'use this when...' or 'instead of...' information. This leaves the agent to infer when this tool is appropriate, which is a notable gap.

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