Fastero

com.fastero由 fastero.com 發布官方發布者v1.0.1更新於 Oct 24, 2026

Query your databases and build dashboards, Streamlit apps and reports your team logs in to.

已驗證Streamable HTTP可網頁執行資料庫雲端與基礎設施資料與分析

概覽

AI 產生的概覽

Fastero 是託管的 MCP 伺服器,讓助手在你的 Fastero 工作區中查詢資料庫,並建立儀表板、Streamlit 應用程式、報表與警示。

可以做什麼
  • 探索資料:列出連線與資料表、執行唯讀 SQL、讀取團隊議定的指標定義。
  • 建立儀表板:儲存查詢並轉成圖表、表格與數字卡片。
  • 分享與排程:邀請同事、發布唯讀公開連結,或依排程寄送附 PDF 的儀表板郵件。
  • 交付應用與程式碼:部署 Streamlit 應用程式,立即或依排程執行 notebook 與 Python 專案。
  • 監控數字:設定警示,在指標跨越門檻時寄送郵件。
適用情境
適合團隊資料存放在 Fastero,或希望助手在聊天用戶端中把資料庫查詢變成共享儀表板、排程報表、警示與小型 Streamlit 應用程式的情境。
執行需求
需要 Fastero 帳號與工作區。透過 streamable HTTP 連線遠端端點並以 OAuth 登入(支援動態用戶端註冊),或使用以 Authorization Bearer 標頭傳送的 API 金鑰。每個工作區有各自的端點位址。
安裝前請注意
助手以你的身分在所选工作區內操作,可建立與編輯儀表板、警示、排程、邀請與應用程式碼,程式碼隨後會被執行。API 金鑰依權限選項限定範圍,最長有效 90 天。在 Fastero 中執行的程式碼會取得你的 Secrets,其輸出內容對助手可見。可在 AI Agent Access 中撤銷連接器與金鑰。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Fastero,將其新增到工作區。
  2. 連接 所需的 憑證。它們會被私密儲存,絕不會在公開頁面上展示。
  3. 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/com.fastero%2Ffastero/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。 客戶端會在首次連接時完成 OAuth 登入。

{
  "mcpServers": {
    "fastero": {
      "type": "http",
      "url": "https://api.fastero.com/api/v1/mcp"
    }
  }
}

README

Fastero MCP server

Query your databases and build dashboards, Streamlit apps and reports your team logs in to, from GitHub Copilot, Claude or another MCP client.

Fastero is a hosted service. This repository holds the connection instructions; the server itself runs at:

https://api.fastero.com/api/v1/mcp

It is listed in the official MCP Registry as com.fastero/fastero.

What you can do

Ask in plain words. Your assistant works inside your Fastero workspace, and what it builds stays there for your team:

  • Explore your data. List your connections and tables, run read-only SQL, and read the definitions your team agreed on, such as what counts as revenue.
  • Build dashboards. Save queries and turn them into charts, tables and number cards on a dashboard your colleagues open with their own login.
  • Share and schedule. Invite teammates, publish a public read-only link to a dashboard, or email a dashboard, with a PDF attached, to people in your workspace every Monday.
  • Ship apps and code. Deploy a Streamlit app behind your Fastero login, and run notebooks and Python projects now or on a schedule, with your secrets injected.
  • Watch the numbers. Get an email when a number crosses a line.

Connect

You need a Fastero account: sign up.

VS Code (GitHub Copilot)

Open the Command Palette and run MCP: Add Server, choose HTTP, and paste the address above. Or add it to your mcp.json:

json
{  "servers": {    "fastero": {      "type": "http",      "url": "https://api.fastero.com/api/v1/mcp"    }  }}

Start the server. When VS Code asks whether it may sign in, choose Allow, then sign in to Fastero, pick a workspace if asked, and approve. In Copilot Chat, use Agent mode.

Claude

Fastero is in Claude's connector directory. In claude.ai, open Customize → Connectors, choose Discover, search for Fastero and choose Connect to Claude. Then sign in to Fastero, pick a workspace if asked, and approve. On a Team or Enterprise plan, an owner adds Fastero under Organization settings → Connectors, and each member then connects.

The connector also works in Claude Desktop, and Claude Code signed in with the same Claude account picks it up automatically.

Other MCP clients

Other clients that support remote MCP servers with OAuth sign-in should work with the same address. Fastero supports dynamic client registration, so there is no app to register first. If your client cannot complete the sign-in, use an API key.

With an API key

For scripts, automation, or a client that cannot sign in: in Fastero, open Settings → Developers → AI Agent Access → Create key, and choose what the key may do. Build dashboards ticks reading, SQL, new connections, and creating and editing. Creating and editing includes alerts, switching existing schedules and triggers on or off, and changing the code of an app or Python project that already runs or is scheduled, which then runs the new code. Running code and setting up schedules; public links, dashboard emails and invites; and Secrets each have their own box. Send the key as Authorization: Bearer <key>.

In VS Code, so the key is asked for once and kept out of the file:

json
{  "inputs": [    { "type": "promptString", "id": "fastero-key", "description": "Fastero API key", "password": true }  ],  "servers": {    "fastero": {      "type": "http",      "url": "https://api.fastero.com/api/v1/mcp",      "headers": { "Authorization": "Bearer ${input:fastero-key}" }    }  }}

In Claude Code signed in with an API key:

claude mcp add --scope user --transport http fastero https://api.fastero.com/api/v1/mcp --header "Authorization: Bearer <key>"

Keys last up to 90 days, and Fastero notifies you in the app 14 and 3 days before one expires.

More than one workspace

Every workspace also has an address of its own, https://api.fastero.com/api/v1/mcp/org/<workspace id>. AI Agent Access shows it with a copy button. In claude.ai, connect the first workspace from the directory, then add each other workspace as a custom connector (Customize → Connectors → Add custom connector) with its own address, named after the workspace.

Security

  • You sign in and approve what the assistant may do. It acts as you, in the one workspace you chose, and never beyond what your own account may do there.
  • Connection passwords, Git tokens and Secret values are never returned by any tool. Code you run in Fastero receives your Secrets, and a Streamlit app linked to a connection receives that connection's login, so anything that code prints is visible to the assistant.
  • Your connectors and keys are listed, per workspace, under Settings → Developers → AI Agent Access, and revoking one ends it at once.

Example requests

  • "Show me the tables in my sales database."
  • "Build a dashboard called Sales overview: revenue and orders for the last month, revenue by month for the last two years, and the ten products with the most revenue."
  • "Email this dashboard to me every Monday at 8am London time."
  • "Deploy the Streamlit app in this folder to Fastero and give me the link."
  • "Alert me when daily signups drop below 20."

Help

Questions and feedback: fastero.com/contact.

來源:README.md,提交 863985f

工具

0
該伺服器需要登入,因此 SourceWeft 無法在沒有帳號的情況下讀取其工具,也未收錄。已連接帳號所能看到的內容可能因帳號及其權限而異。

版本歷史

2
  1. v1.0.1最新Oct 15, 2026
  2. v1.0.0Oct 13, 2026

評分

已滿足適用檢查項的 75%

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15 分

    請讓伺服器在無憑證時回應 tools/list,或發布可在沙箱中無憑證啟動的軟體包。

  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。