MCP server for Chestniy ZNAK — product marking verification, CRPT (Russia)
Server has 5 tools with complete input schemas and descriptions in Russian. All tools are READ_ONLY and properly registered with Zod schemas. However, descriptions are domain-specific (Russian product marking) and lack LLM-optimized guidance on WHEN to use each tool vs. alternatives. Output schemas are defined but tolerant (nullable fields). Parameter descriptions are adequate but could be more explicit about constraints and error cases. No tool annotations (readOnlyHint, etc.) despite all being read-only. Error handling returns structured JSON with isError flag, which is good, but recovery guidance is minimal.
Пакетная проверка до 50 кодов маркировки за один запрос.
Проверка подлинности товара по коду маркировки Честный ЗНАК.
Информация о CIS (коде идентификации) из True API. Требует CHESTNYZNAK_TOKEN.
Подробная информация о товаре по коду маркировки: название, группа, производитель, владелец.
Информация о товаре по GTIN из True API. Требует CHESTNYZNAK_TOKEN. (Поиск по названию — Нацкаталог, вне API.)
No tool annotations despite all tools being read-only. Missing readOnlyHint metadata that would signal to agents these are safe, non-destructive operations.
Descriptions lack LLM-optimized guidance on tool selection. No explanation of WHEN to use check_marking_code vs. get_product_info vs. check_batch, or when search_products/get_cis_info require CHESTNYZNAK_TOKEN.
Output schemas use tolerant Zod definitions with nullable fields, but do not document what each field means or when it will be null. LLMs cannot infer field semantics from schema alone.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
Error handling returns isError=true with JSON error message, but does not provide recovery guidance. E.g., if CHESTNYZNAK_TOKEN is missing, error should suggest 'Set CHESTNYZNAK_TOKEN environment variable' rather than raw error text.
Parameter descriptions in Russian may not be parsed correctly by English-language LLMs. Descriptions like 'Код маркировки товара' lack English context and constraint details (e.g., expected length, format).