Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@weather_mcp what's the 7-day forecast for New York?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Учебный проект: MCP-сервер ходит в публичный API Open-Meteo, консольный клиент спрашивает город (кириллицей или латиницей) и печатает прогноз на 7 дней.
MCP (Model Context Protocol) — открытый стандарт Anthropic (конец 2024). Коротко для собеседования: это «USB-C для ИИ». Один протокол, чтобы подключать к модели внешние системы (API, файлы, БД) без самописного плагина под каждый чат.
Три роли:
Host — приложение, в котором живёт модель (Cursor, Claude Desktop) или, как здесь, наше CLI.
Client — сторона, которая говорит с сервером по протоколу: handshake, список инструментов, вызов.
Server — процесс, который отдаёт инструменты. Он не чат и не модель, а розетка с функциями.
Три примитива протокола (в этом проекте только первый):
Tools — функции с JSON Schema аргументов. Методы tools/list и tools/call. У нас это get_forecast(city).
Resources — данные «как файлы» по URI. Не используем.
Prompts — шаблоны промптов. Не используем.
Связь — JSON-RPC 2.0. Локально идёт по stdio: клиент запускает сервер как subprocess и пишет JSON-строки в его stdin. Поэтому сервер нельзя логировать через print() — это попадёт в тот же поток, что и протокол, и сломает его. Логи только в stderr (logging).
«Классический» хост — Cursor: модель сама решает, какой tool вызвать. Здесь хост — client.py: город вводит человек, клиент явно вызывает get_forecast. Протокол тот же, LLM нет. Так проще увидеть handshake, list_tools и call_tool без «магии модели».
REST — как сервер разговаривает с Open-Meteo. MCP — как ИИ-приложение разговаривает с вашим сервером: единый каталог инструментов, схемы аргументов, один транспорт. Сервер пишете один раз — его могут вызвать Cursor, Claude Desktop или этот CLI.
Спецификация: modelcontextprotocol.io.
Related MCP server: Weather MCP Server
Вы запускаете клиент и вводите город (Москва или Moscow).
Клиент поднимает MCP-сервер по stdio и вызывает tool get_forecast.
Сервер ищет координаты в Geocoding API (кириллица → language=ru, иначе en) и берёт 7 дней из Forecast API.
Клиент печатает таблицу: дата, день недели, погода, мин/макс, осадки, ветер.
Ключ API не нужен.
Нужны Python 3.10+ и uv.
cd Checking_weather_MCP
uv sync
uv run python -m weather_mcp.clientЕсли команда uv не находится, тот же CLI ставится через pip: py -m pip install uv, дальше py -m uv sync и py -m uv run python -m weather_mcp.client.
Появится приглашение Город:. После ввода — таблица на неделю.
Сервер отдельно обычно не запускают: без клиента он просто ждёт JSON-RPC на stdin. Если нужно вручную:
uv run python -m weather_mcp.server7-дневный прогноз по названию города (кириллица или латиница).
| Name | Required | Description | Default |
|---|---|---|---|
| city | Yes |
| Name | Required | Description |
|---|---|---|
| city | Yes | |
| days | No | |
| country | Yes | |
| latitude | Yes | |
| timezone | Yes | |
| longitude | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the behavioral disclosure burden. It does disclose the forecast horizon and accepted input script, but it does not mention possible limitations, error behavior, timezone considerations, or data source. The description is not misleading but is minimal.
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?
A single compact sentence that front-loads the core semantics and the input requirement. There is no redundancy or filler; every part contributes useful information.
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 one-parameter tool with an output schema, the description covers the essential input and output semantics. Minor gaps like units, city ambiguity, or error behavior are not addressed, but the tool's simplicity and the presence of an output schema keep these gaps acceptable.
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 0%, so the description must compensate. It adds meaning by clarifying that the city value is a name (rather than an ID) and explicitly supports both Cyrillic and Latin script. This goes beyond the bare string type in the schema, though it doesn't specify additional formatting or disambiguation details.
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 states a clear verb and resource: it returns a 7-day forecast for a given city name. It is specific and unambiguous, and the additional script hint (Cyrillic or Latin) further sharpens the purpose.
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?
There are no sibling tools or explicit when/when-not guidance, but the description implies the intended use: whenever a 7-day city forecast is needed. This is adequate for a simple single-purpose tool but leaves the agent to infer the context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.1.0get_forecastOnly one tool exists, so there is no possibility of confusion or overlap. The single tool's purpose is clearly defined as a 7-day forecast by city name.
The tool name get_forecast follows a clear verb_noun convention that would be consistent with a broader weather toolset. However, with only one tool, the naming pattern is not fully demonstrable.
A single tool is far too few for a weather server, which typically requires current conditions, geocoding, unit preferences, and alerts. This feels like an extreme under-scoping of the domain.
The tool surface is severely incomplete for a weather service. It only provides a 7-day forecast, leaving out current weather, location search, weather alerts, and other common weather data operations.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Free, keyless real-time weather and 7-day forecasts for any city worldwide.
Real-time weather conditions and multi-day forecasts via Open-Meteo — free, no API key required
Current weather and forecasts for any coordinates, backed b… — paid per call (x402/credits), 1 tools
Global weather via Open-Meteo: forecast, ERA5 archive, marine, air quality, geocoding, elevation.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/AronSoldok/Checking_weather_MCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server