Tern EV-Charging Analytics

Tern EV-Charging Analytics: загрузка и спрос на зарядных станциях, сравнение по коридорам

От сообществаСбоиНужен API-ключГлобальныйБесплатноТолько чтение

Что умеет

  • Искать зарядные станции и коннекторы рядом
  • Считать загрузку станций и пиковую насыщенность
  • Сравнивать станции и коридоры между собой
  • Смотреть динамику сессий по месяцам и дням
  • Оценивать контекст POI и дорог вокруг станции

Какие данные видит

Ассистент обращается к данным о зарядных станциях через API-ключ, доступ ограничен правами ключа.

Нужен ли аккаунт

Нужен API-ключ из настроек сервиса

MCP-сервер отдаёт аналитику по сети зарядных станций для электромобилей: загрузку станций, пиковые часы, сравнение площадок и транспортных коридоров. Подходит для оценки эффективности размещения станций и планирования сети зарядки.

Как подключить

Как подключить Tern EV-Charging Analytics к Claude

Адрес сервера, который нужно вставить в настройках Claude:

https://tern-mcp-537572042041.europe-west4.run.app/mcp

  1. Откройте Claude (сайт claude.ai или приложение для компьютера).
  2. Перейдите в Настройки → Коннекторы (в английском интерфейсе: Settings → Connectors).
  3. Нажмите Добавить пользовательский коннектор (Add custom connector).
  4. Вставьте адрес сервера, скопированный ниже, в поле URL удалённого MCP-сервера и нажмите Добавить.
  5. В чате нажмите + → Коннекторы и включите новый коннектор.

Пользовательские коннекторы доступны на тарифах Free, Pro, Max, Team и Enterprise. На Free можно добавить только один. В Team и Enterprise коннектор сначала добавляет владелец организации в Настройки организации → Коннекторы.

Авторизация

После нажатия Добавить откроется окно входа в сервис. Войдите под своим аккаунтом и подтвердите доступ. Пароль Claude не видит.

Как подключить Tern EV-Charging Analytics к ChatGPT

Адрес сервера, который нужно вставить в настройках ChatGPT:

https://tern-mcp-537572042041.europe-west4.run.app/mcp

  1. Откройте ChatGPT в браузере (chatgpt.com). Нужен тариф Plus, Pro, Business, Enterprise или Edu.
  2. Один раз включите режим разработчика: Настройки → Приложения → Дополнительные настройки → Режим разработчика (Settings → Apps → Advanced settings → Developer mode).
  3. Откройте Настройки → Коннекторы и нажмите Создать (Create).
  4. Заполните форму: Название (любое), Описание (одна фраза, что делает сервис), URL MCP-сервера (скопируйте ниже).
  5. В поле Аутентификация выберите OAuth, если сервис требует вход, иначе Нет. Нажмите Создать.
  6. В новом чате откройте + → Приложения/Коннекторы и включите коннектор.

Название раздела у OpenAI менялось (Connectors → Apps/Plugins). Если пункт называется иначе, ищите «режим разработчика» в настройках. В рабочих пространствах Business/Enterprise администратор должен разрешить пользовательские коннекторы.

Авторизация

При первом использовании ChatGPT откроет окно входа в сервис. Войдите и подтвердите доступ.

Как подключить Tern EV-Charging Analytics к Cursor

Адрес сервера, который нужно вставить в настройках Cursor:

https://tern-mcp-537572042041.europe-west4.run.app/mcp

Быстрый способ: нажмите кнопку Открыть в Cursor ниже, Cursor сам предложит добавить сервер.

Вручную:

  1. В Cursor откройте Settings → Cursor Settings → MCP и нажмите Add new global MCP server.
  2. Вставьте JSON, скопированный ниже, в открывшийся файл ~/.cursor/mcp.json (для одного проекта: .cursor/mcp.json в корне проекта).
  3. Сохраните файл. Сервер появится в списке MCP; при необходимости пройдите авторизацию по кнопке рядом с ним.

Авторизация

Если сервис требует вход, рядом с сервером в списке MCP появится кнопка авторизации.

Список инструментов сервера (26)

Технические названия из tools/list. Нужны только разработчикам.

search_charging_stationsFind charging stations in one country by structured filter — id, title, address, operator. The reply is the API payload inline: station ids, titles, addresses and operators, which is what the claim tools take as site_id and station_id.
nearby_charging_stationsList the charging stations within a radius of a point, nearest first. The radius is in kilometres and at most 10; the reply is the API payload inline. This is how a site's competitors are found before any claim is read.
list_evsesList the individual EVSEs (connectors) of up to 20 stations, with their capacity in kW. Stall counts and capacity bands come from here rather than from a station record.
aggregate_charging_utilizationStation-level utilization aggregated over a GeoJSON area and a date window of at most 92 days, optionally filtered to a capacity band (default 150 kW = HPC; pass 100 for fast charging). The reply is one row per station, so it reads an area without pulling any station's daily rows.
aggregate_evse_utilizationEVSE-level utilization aggregated over a GeoJSON area and a date window of at most 92 days, optionally filtered to a capacity band. The per-connector counterpart of aggregate_charging_utilization.
area_benchmarkHow hard is each EVSE at this site working compared with the area around it? Reads the area_benchmark claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
avg_session_minutesHow long does a session at this site actually last? Reads the avg_session_minutes claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
busy_evse_hoursHow many EVSE-hours at this station carried charging on the drilled day, and how many of its EVSEs were charging at once in the busiest hour? Reads the busy_evse_hours claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
corridor_benchmarkHow does this site's demand per EVSE-day compare with the corridor drawn around it? Reads the corridor_benchmark claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
corridor_comparisonHow does this site compare with the named stations around it — distance, EVSE count, max kW, and sessions per EVSE-day each? Reads the corridor_comparison claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
daily_spikeWhich days ran far above this site's normal, and by how much? Reads the daily_spike claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
day_vs_similarWas one particular day unusual for the day of week it fell on? Reads the day_vs_similar claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
demand_share_multipleHow much of the area's demand does this one site take, said as a multiple of the mean station? Reads the demand_share_multiple claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
estate_coverageWhich estate sites reported on every day of the window, which are partial or young, and how full is the window's final day? Reads the estate_coverage claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
estate_monthly_seriesDid demand per EVSE across the estate move month over month, and did each focus site move with it? Reads the estate_monthly_series claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
hourly_spikeWhen in the day does demand at this site peak? Reads the hourly_spike claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
monthly_sessions_per_stall_dayIs demand per EVSE rising or falling month over month? Reads the monthly_sessions_per_stall_day claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
peak_hour_co_saturationWhen the site reaches its busiest hour of the day, do the stations around it reach their own busiest hours at the same time? Reads the peak_hour_co_saturation claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
poi_contextWhat kind of place is this site sitting in? Reads the poi_context claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading. places carries the rows this claim reads, because they are transcribed by the caller rather than fetched.
rank_gapWhere does this site sit in its estate's ranking, and how far is it from the estate median and the estate leader? Reads the rank_gap claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
road_contextWhat road does this site sit on, and what class is it? Reads the road_context claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
sessions_per_stall_dayHow hard is each EVSE at this site working per day? Reads the sessions_per_stall_day claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
site_profileWhat is this site — how many EVSEs, and how close is the nearest station? Reads the site_profile claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
site_rankingHow do the estate's sites rank on sessions per EVSE-day over one window, and how far is the leader above the estate median? Reads the site_ranking claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
total_charging_minutesHow many minutes of charging did this site actually deliver over the window? Reads the total_charging_minutes claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.
total_sessionsHow much charging demand did this site see over the window? Reads the total_sessions claim for one site and returns only its result: the status the claim decided from its own observations (computed or insufficient), the typed result, and the display a deliverable would print. The observations behind it are pulled into a scratch directory that is deleted before the call returns, so the rows never enter the conversation and nothing is left on the server to read back later — this call is the whole reading.