V-Lab MCP

V-Lab MCP: волатильность, SRISK, CRISK и другие метрики риска от NYU Stern через ассистента

От сообществаРаботаетБез входаГлобальныйБесплатноМожет изменять данные

Что умеет

  • Искать активы по тикеру, названию или идентификатору
  • Получать волатильность по странам, секторам и отдельным активам
  • Смотреть системный и климатический риск компаний (SRISK, CRISK)
  • Анализировать ликвидность и показатели VaR
  • Изучать климатические бенчмарки и корреляции факторов

Какие данные видит

Сервер отдаёт открытые академические данные V-Lab; для подключения нужен бесплатный аккаунт V-Lab и авторизация по OAuth 2.1 с PKCE.

Нужен ли аккаунт

Не нужен: сервер работает без входа

V-Lab MCP отдаёт данные Института волатильности и риска NYU Stern: волатильность по странам и секторам, системный риск SRISK, климатический риск CRISK, показатель COVOL, ликвидность ILLIQ и оценки VaR. Все данные опираются на опубликованные академические исследования и доступны через Claude, Claude.ai или MCP Inspector по OAuth 2.1.

Как подключить

Как подключить V-Lab MCP к Claude

Адрес сервера, который нужно вставить в настройках Claude:

https://vlab.run.tools

  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 V-Lab MCP к ChatGPT

Адрес сервера, который нужно вставить в настройках ChatGPT:

https://vlab.run.tools

  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 V-Lab MCP к Cursor

Адрес сервера, который нужно вставить в настройках Cursor:

https://vlab.run.tools

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 (45)

Технические названия из tools/list. Нужны только разработчикам.

server.infoGet information about the V-Lab MCP server including capabilities, status, and available features
search.assetsResolve a financial asset by ticker, name, FIGI, SEDOL, or GVKEY. Returns compact results (ticker, name, active, last_result_date). Pass `include_analyses:true` for the full analyses catalog.
search.datasetsSearch for datasets and discover their constituent assets and available analyses. Returns datasets matching the query along with paginated asset lists and analysis information.
search.asset_in_datasetCheck if a specific asset is a constituent of a dataset and retrieve available analyses for that combination.
search.analysesSearch for analyses by application, model, or memo. Returns flat results with full context for each analysis instance.
volatility.getAnnualized volatility time series for an asset (GARCH-family models).
liquidity.getGet liquidity time series data for an asset. Returns illiquidity measures from ILLIQ models. SMEM and MFMEM models also include long-term trend data.
volatility.paramsLatest estimated GARCH-family parameters for an asset (omega, alpha, beta, gamma, etc.), each with its standard error, plus model persistence and half-life. One parameter set per analysis — the most recent successful fit; there is no parameter history.
volatility.forecastDynamically-generated multi-step volatility forecast (the cumulative-average annualized term structure) at user-defined horizons. Computed by iterating the fitted model forward — matching the V-Lab website's forecast chart — so any steps-ahead can be requested, not just the stored 1d/1w/1m/6m/1y points. Multi-factor models (MF2-GARCH) are interpolated across the stored summary forecasts.
liquidity.paramsLatest estimated liquidity (MEM / ILLIQ-family) parameters for an asset, each with its standard error, plus model persistence and half-life. Vector parameters (e.g. spline knot coefficients) are returned as arrays. One parameter set per analysis — the most recent successful fit; there is no parameter history.
liquidity.forecastDynamically-generated multi-step illiquidity (ILLIQ) forecast (the cumulative-average term structure) at user-defined horizons. Computed by iterating the fitted model forward — matching the V-Lab website's forecast chart — so any steps-ahead can be requested. Multi-factor models (ILLIQ-MFMEM) are interpolated across the stored summary forecasts.
liquidity.illiq_compositeGet the ILLIQ Composite market-wide liquidity index. This time series shows aggregate market liquidity conditions based on market-cap weighted average of individual ILLIQ measures. Higher values indicate less liquid markets.
liquidity.list_change_countriesList countries that have aggregated sector-change data — the input universe for `liquidity.changes`. Returns each country with its asset count. Note: this is not the universe of every country with ILLIQ coverage; per-asset tools like `liquidity.get` and `liquidity.movers` work on a broader universe.
liquidity.changesGet sector-level liquidity statistics for a country. Shows which GICS sectors are experiencing liquidity stress (deteriorating) or improvement.
liquidity.moversGet assets with highest illiquidity ("hot") or fastest deteriorating liquidity ("heating"). Use to identify liquidity-stressed assets.
volatility.global_mapGet relative volatility percentiles for all countries. Supports single-date snapshots OR time series with `start_date`/`end_date`. Data available from 1990 to present. Drill down: `volatility.country.get` (sectors) or `volatility.country.summary` (key indices).
volatility.country.getGet GICS sector/industry volatility breakdown for a specific country. Without `industry` param: returns sectors. With `industry` param (e.g. `"20"`): returns sub-industries. Drill down: `volatility.country.industries` (individual assets within an industry).
volatility.country.summaryGet market summary for a country showing key indices with current volatility levels and changes, grouped by market type (Equities, Currencies, etc.). Drill down: `volatility.get` (full time series for any asset).
volatility.country.industriesGet individual assets within a GICS industry for a country, with relative volatility percentiles and levels. Use after `volatility.country.get` to drill down from sector/industry to individual assets. Part of the volatility hierarchy.
climate_benchmarks.listList available climate risk benchmarks. These are V-Lab's proprietary climate factor portfolios including Stranded Assets, Emissions Factor, Oil Beta Factor, and Subsidy Factor.
climate_benchmarks.returnsGet returns time series for a climate benchmark. Returns cumulative returns by default (base 100), or daily returns if cumulative=false.
climate_benchmarks.volatilityGet annualized volatility time series for a climate benchmark.
climate_benchmarks.correlationsGet correlation matrix between all available climate benchmarks.
climate.categoriesList the climate-fund category taxonomy — the ESG / low-carbon "groups" V-Lab classifies funds into (e.g. Broad ESG, Low Carbon, Sustainable Sector, Fossil Fuel Free, Carbon Trading) — with the number of active funds in each. Use it to discover the valid values for `climate.assets`'s `categories` filter. A fund may belong to more than one category.
climate.assetsList V-Lab's climate-fund universe — ESG / low-carbon ETFs and mutual funds — each with its categories and trailing 1-year analytics (annualized volatility and return). Optionally restrict to one or more categories. Distinct from `climate_benchmarks.*`, which are factor portfolios, not investable funds. Use `climate.categories` to discover the category names.
srisk.listList available SRISK (systemic risk) analyses with regions and coverage. Part of the SRISK hierarchy: `srisk.list` → `srisk.ranking` → `srisk.country` → `srisk.firm`.
srisk.rankingGet top entities ranked by SRISK (systemic risk). Can rank firms (default), countries, regions (continents), or markets. Includes concentration metrics (HHI, top-N share). Part of the SRISK hierarchy: `srisk.list` → `srisk.ranking` → `srisk.country` → `srisk.firm`. Data note: SRISK relies on balance sheet data which may arrive with a lag; values for recent quarters may be revised as updates are received from data providers.
srisk.countryGet SRISK data for a specific country. Returns either firm rankings (default) or country-level time series with `time_series=true`. Part of the SRISK hierarchy: `srisk.list` → `srisk.ranking` → `srisk.country` → `srisk.firm`.
srisk.firmGet SRISK time series for a specific firm. Part of the SRISK hierarchy: `srisk.list` → `srisk.ranking` → `srisk.country` → `srisk.firm`.
srisk.moversGet firms with largest SRISK changes over a period. Returns top increases and/or decreases with change attribution (debt/equity/risk breakdown). Part of the SRISK hierarchy for identifying risk trends.
crisk.listList available CRISK (climate risk) analyses with coverage. CRISK is the climate-stress analog of SRISK, measuring expected capital shortfall under a climate-risk factor shock. Part of the CRISK hierarchy: `crisk.list` → `crisk.ranking` → `crisk.country` → `crisk.firm`.
crisk.rankingGet top entities ranked by CRISK (climate-stress capital shortfall). Can rank firms (default), countries, regions (continents), or markets. Includes concentration metrics (HHI, top-N share). Part of the CRISK hierarchy: `crisk.list` → `crisk.ranking` → `crisk.country` → `crisk.firm`. Data note: CRISK relies on balance sheet data which may arrive with a lag; values for recent quarters may be revised as updates are received from data providers.
crisk.countryGet CRISK data for a specific country. Returns either firm rankings (default) or country-level time series with `time_series=true`. Part of the CRISK hierarchy: `crisk.list` → `crisk.ranking` → `crisk.country` → `crisk.firm`.
crisk.firmGet CRISK time series for a specific firm. CRISK is computed from components (climate beta, market cap, book assets, book equity) using a 50% climate stress shock. Part of the CRISK hierarchy: `crisk.list` → `crisk.ranking` → `crisk.country` → `crisk.firm`.
crisk.moversGet firms with largest CRISK changes over a period. Returns top increases and/or decreases with change attribution (debt/equity/risk breakdown). Part of the CRISK hierarchy for identifying climate risk trends.
covol.listList available COVOL (common volatility) analyses with coverage. COVOL extracts a synchronized-stress factor from the cross-section of daily returns within each dataset. Each analysis covers a different universe (e.g., country ETFs, asset classes, commodities). Part of the COVOL hierarchy: `covol.list` → `covol.summary` → `covol.get`.
covol.summaryCOVOL stress snapshot: composite PC1 plus per-analysis levels across every available COVOL analysis.
covol.compositeGet Composite COVOL Index time series. Returns the systematic stress component (PC1) with daily factor loadings showing how each analysis contributes to market-wide stress.
covol.cavGet COVOL-Adjusted Volatility (CAV) time series for each analysis and aggregate (ACAV). CAV is the annualized risk of the common factor within each analysis. ACAV combines all four into a single portfolio-level risk metric.
covol.getGet COVOL Index time series for a specific analysis. Use `covol.list` to discover the available analyses and their memo strings.
covol.loadingsGet asset loadings (factor sensitivities) for a COVOL analysis. Shows how much each asset contributes to common volatility, both equal-weighted and variance-weighted.
covol.eventsGet top COVOL events (highest stress dates) for an analysis with z-scores and event descriptions. Events include major market crashes, geopolitical events, and policy shocks.
lrvar.summaryCurrent Long-Run VaR across all horizons and percentiles (decimals, e.g., -0.1435 = 14.35% loss). Returns LRGJRF (return-series) and/or LRGJROF (options-augmented) where available.
lrvar.getLong-Run VaR time series for an analysis (decimals, e.g., -0.1435 = 14.35% loss). Horizons: 30d (tactical) or 365d (strategic). Percentiles: 1 or 5.
feedback.submitReport feedback on V-Lab MCP to help improve the server. TWO SHAPES ARE SUPPORTED: 1. Tool rating — pass tool_name + helpful (+ optional message): { tool_name: "volatility.get", helpful: true, message: "Data was clear and correct" } { tool_name: "srisk.ranking", helpful: false, message: "aggregate_by=firm returned a DB error" } 2. General feedback — pass category + message (no tool_name required): { category: "bug_report", message: "covol.cav is listed in server_info but has no tool schema" } { category: "feature_request", message: "Add a tool that compares volatility across multiple assets" } { category: "data_request", message: "Need coverage of Brazilian small-cap banks" } { category: "general", message: "Overall very useful for market-state summaries" } WHEN TO USE: - After completing a task, if you have meaningful input on a specific tool (helpful=true/false). - Any time to file a bug, request a feature, request data coverage, or share general feedback — no tool name required.