EPW.tools

io.github.carlobianchi89由 GitHub 帳號 carlobianchi89 發布v0.5.3更新於 Nov 4, 2026

EnergyPlus weather files (EPW, DDY, STAT) and design conditions for any site, OpenStudio-MCP ready.

已驗證Streamable HTTP可網頁執行位置與生活資料與分析

概覽

AI 產生的概覽

EPW.tools 可為任意地點產生 EnergyPlus 氣象檔（EPW、DDY、STAT）、設計條件與預報，供建築模擬使用。

可以做什麼
  • 查詢地點附近的氣象站以及 ASHRAE 設計條件
  • 依 URL 分析公開 EPW 檔，提供統計摘要與資料警告
  • 啟動實測年、典型年（TMY）、CMIP6 未來氣候以及 1 至 15 天預報氣象檔的產生工作
  • 列出可用於未來氣候轉換的 CMIP6 全球與區域模型
  • 查詢工作狀態並取得已完成檔案的下载連結
適用情境
適合在協助建築能耗模擬時使用：需要某地點的氣象檔或設計條件，包括典型年、未來氣候情境或短期預報，且希望不必離開對話就能完成。
執行需求
可使用託管的遠端端點，或透過 uvx 從含 MCP 擴充的 epwtools 套件在本機執行。產生類工具需要以 api_key 傳入 EPWTOOLS_API_KEY；讀取類工具免費且不需金鑰。免費帳號提供少量初始額度。
安裝前請注意
產生類工具會消耗額度：每個檔案 1 點，未來氣候 3 點，街道級城市資料另加 2 點；預報檔案 7 天內免費。API 金鑰會傳給伺服器，下载連結中帶有金鑰，應視為機密。工作為非同步執行，需要輪詢結果。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 EPW.tools，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.carlobianchi89%2Fepwtools/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "epw-tools": {
      "type": "http",
      "url": "https://epw.tools/api/mcp"
    }
  }
}

README

epwtools

[carlobianchi89/epwtools-python MCP server]

Python client, command-line tool and MCP server for EPW.tools: EnergyPlus weather files (EPW, with DDY design days and STAT summaries) for any location on Earth.

  • Actual years from 1990 to now, for any coordinates, with or without a nearby weather station
  • Typical years (TMY), extreme and heatwave years
  • Future climate for 2030 to 2100 under SSP scenarios
  • Design conditions (ASHRAE method) and short-term forecast weather files

The weather data are built and validated by EPW.tools (methodology, validation); this package is the open-source way to call them from Python, the terminal or an AI assistant.

Install

bash
pip install epwtools            # client + CLIpip install 'epwtools[mcp]'     # + MCP server for AI assistants

Get a free API key at epw.tools (API tab); the first files are free.

bash
export EPWTOOLS_API_KEY=your-key

Python

python
from epwtools import EPWClient
client = EPWClient()                       # reads EPWTOOLS_API_KEY
# actual year 2023 for a site in Denverepw = client.generate(lat=39.74, lon=-104.99, year=2023, name="Denver", out_dir="weather")
# typical year built from 2005-2024tmy = client.tmy(lat=45.46, lon=9.19, start=2005, end=2024, name="Milan", out_dir="weather")
# design conditions (ASHRAE method)dc = client.design_conditions(lat=51.51, lon=-0.13)

Command line

bash
epwtools configure                          # store your keyepwtools generate --lat 39.74 --lon -104.99 --year 2023 --name Denverepwtools tmy --lat 45.46 --lon 9.19 --start 2005 --end 2024 --name Milanepwtools future --lat 40.71 --lon -74.01 --scenario ssp245 --horizon 2050 --name NYCepwtools --help

AI assistants (MCP)

Let Claude, ChatGPT and other MCP-capable assistants make weather files for you:

bash
uvx --from 'epwtools[mcp]' epw-tools-mcp

or use the hosted server, listed in the official MCP Registry. Tools include generate_epw, generate_tmy, generate_future_climate, generate_forecast, get_design_conditions and get_nearest_stations. Setup guide: epw.tools/ai-weather-files.

Also on EPW.tools (web)

EPW to CSV / Excel converter · Manual J design temperatures by address · Construction weather days · Data-centre site climate · API licence for software vendors

Licence

The client, CLI and MCP server in this repository are MIT-licensed. Use of the EPW.tools service and its data is governed by the EPW.tools terms of service (linked in the footer of epw.tools).

來源：README.md，提交 f1e0fc9

工具

9

get_nearest_stations

Nearest weather stations

Nearest weather stations to a point (name, WMO id, distance, years of record). Free, no key.

輸入參數nlat必填lon必填

get_design_conditions

Design conditions (ASHRAE method or EPW.tools Enhanced)

Heating and cooling design conditions for any site, free, in one of two modes that are never mixed. mode="ashrae_method": the ASHRAE Handbook Ch. 14 method replicated at the nearest ASHRAE station for the chosen edition (2025, 2021, 2017, 2013, 2009, 2005); validated about 0.25 degC from the published tables. mode="enhanced": EPW.tools Enhanced, not the ASHRAE method: values at the site itself for today's climate with a tested 90 % range. The DDY file for Enhanced costs 1 credit through the web or the API.

輸入參數lat必填lon必填modeeditionlocation_namecurrent_climate

list_cmip6_models

CMIP6 models for future climate

The models available for future-climate morphing: `models` are global CMIP6 models (NASA NEX-GDDP, 25 km); `regional` are EURO-CORDEX pairs (12.5 km, Europe only, 27-72 N / 25 W-45 E) usable as `model` ids. Free.

無需輸入參數。

analyze_epw

Analyze an EPW by URL

Summary statistics of any public EPW by URL (OneBuilding, EnergyPlus, a shared file): location header, dry-bulb mean/min/max, annual global horizontal irradiation, data range warnings. Free, no key. 10 MB max.

輸入參數url必填

generate_epw

Generate actual-year EPW (AMY)

Start an actual-year (AMY) EPW + DDY + STAT for any point, 1990 to last year. 1 credit; +2 with urban_street_level (53 US cities: hourly street-level air temperature from citizen stations). Returns a job_id; poll with get_job_result.

輸入參數lat必填lon必填year必填api_key必填dc_editiondata_sourcelocation_nameurban_street_level

generate_tmy

Generate typical year (TMY)

Start a typical meteorological year from a chosen period (at least 5 years; 10-20 is usual). 1 credit.

輸入參數lat必填lon必填api_key必填end_year必填dc_editionstart_year必填location_name

generate_future_climate

Generate future-climate EPW (CMIP6)

Start a CMIP6-morphed future EPW (Belcher 2005). scenario: ssp126, ssp245, ssp370, ssp585; horizon: any decade 2030-2100; model from list_cmip6_models. 3 credits.

輸入參數lat必填lon必填modelapi_key必填horizonscenariobase_yearlocation_name

generate_forecast

Generate forecast EPW

Start a forecast EPW for the next 1-15 days (operational sims, MPC, demand response). Up to 7 days free.

輸入參數lat必填lon必填daysapi_key必填location_name

get_job_result

Job status and download links

Status of a job started by a generate_* tool; when complete, the download URLs (they carry the key).

輸入參數job_id必填api_key必填

版本歷史

4
  1. v0.5.3最新Nov 4, 2026
  2. v0.5.2Oct 26, 2026
  3. v0.5.1Oct 17, 2026
  4. v0.4.0Sep 29, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分