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

The annotations already declare readOnlyHint=true and idempotentHint=true, and the description reinforces both: it returns modified HTML without publishing, and states 'Преобразования идемпотентны: повторный вызов ничего не меняет и честно отвечает changed=false.' The description also reveals the changed=false contract and the empty-change-list semantics ('Пустой список изменений — не ошибка'), adding valuable behavioral detail beyond what the annotations convey. It doesn't contradict annotations.

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?

Well-structured: each paragraph earns its place — purpose, idempotency contract, exclusion list, and measurement guidance. Uses bold ⚠️ callout for the critical exclusion section. Informative without padding; every sentence conveys a distinct fact about tool behavior or usage.

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 2-parameter tool with full schema coverage, an output schema, and rich annotations, the description comprehensively covers the workflow: it explains what the tool returns, how to complete the loop (publish_landing), what it won't fix, and how to interpret performance results. The interplay with check_performance, site_issues, and publish_landing is fully specified, which is complete for the operational context.

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 coverage is 100%, so the schema already documents both parameters (html = file contents, fix = enum with per-value explanations). The description adds context about what each fix does (e.g., lazy_images skips the first image due to LCP) by referencing the schema's fix description. The description doesn't add new parameter-level semantics beyond the schema, which is the baseline for full coverage.

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 starts with a crisp statement: 'Одно названное преобразование разметки — с предсказуемым результатом.' It names the specific verb (refactor/apply transformation) and resource (site markup), and enumerates exactly what transformations exist (meta, lazy_images). The strong negative framing — 'НИЧЕГО НЕ ПУБЛИКУЕТ' — distinguishes it clearly from publish_landing and site_issues siblings.

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?

Explicit when-to-use guidance: use it for named markup transformations, and explicitly directs the agent to call publish_landing after human approval. It also gives a clear exclusion list — fonts from third-party hosts, webp re-encoding, inline-style parsing — and points to site_issues for those. It even explains when NOT to report success (measurement noise <3 points via check_performance).

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