Imageat

io.github.incendiesv0.4.0更新於 Oct 6, 2026

Generate and edit images and videos with imageat.

已驗證Streamable HTTP可網頁執行AI & MLMedia & Design

概覽

AI 產生的概覽

讓助理透過 ImageAT 生成與編輯圖片和影片，回傳 CDN 連結並可查詢帳戶額度。

功能
把 ImageAT 的生成與編輯功能包裝成 MCP 工具。工具包含用於文生圖與圖生圖的 imageat_generate_image、用於文生影片與圖生影片的 imageat_generate_video，以及查詢目前額度餘額的 imageat_check_credits。每個編輯功能會註冊一個工具，例如 remove-background、object-eraser、relight、virtual-try-on、city-teleport 與 ai-edit-pro，並在啟動時即時取得；若無法取得該功能目錄，則改為註冊一個帶 feature 參數的通用 imageat_edit_image 工具。產生的媒體以 CDN 圖片或 mp4 連結回傳。
適用情境
適合需要助理依需求產生或修改視覺素材的情境，例如依提示詞生成圖片或短影片、去除背景、擦除物件、重新打光或進行虛擬試穿。也適合結果以託管連結而非本機檔案形式使用的工作流程。
執行需求
需要 ImageAT 帳戶，以及在 ImageAT 專案頁面建立的 API 金鑰。本機 npm 套件 @imageat/mcp（需 Node.js，透過 npx 執行）以環境變數 IMAGEAT_API_KEY 提供金鑰；遠端 Streamable HTTP 端點則以 Authorization Bearer 標頭提供。選用變數包括 IMAGEAT_BASE_URL、PORT、MCP_OAUTH_SECRET 與 MCP_PUBLIC_URL。需要能連線至 ImageAT 的網路。
安裝前請注意
生成與編輯會消耗 ImageAT 帳戶額度，也就是工具呼叫會產生費用。API 金鑰是有效憑證（iat_live_ 前綴），須妥善保密；瀏覽器用戶端使用 OAuth 授權流程貼上金鑰，遠端部署應設定 MCP_OAUTH_SECRET，否則重新啟動後已簽發的權杖會失效。提示詞與輸入圖片會傳送到 ImageAT 伺服器，結果託管於其 CDN。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Imageat，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。

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "imageat": {
      "type": "http",
      "url": "https://mcp.imageat.com/mcp"
    }
  }
}

README

@imageat/mcp

Model Context Protocol (MCP) server for ImageAT. Gives any MCP client (Claude Desktop, Cursor, etc.) tools to generate images, generate video, and run ImageAT's image-edit features — billed against your ImageAT account credits.

Tools

ToolWhat it does
imageat_generate_imageText-to-image / image-to-image. Returns CDN image URL(s).
imageat_generate_videoText-to-video / image-to-video. Returns a CDN mp4 URL.
imageat_check_creditsCurrent credit balance.
imageat_edit_<feature>One tool per edit feature, fetched live at startup — e.g. imageat_edit_remove-background, imageat_edit_object-eraser, imageat_edit_relight, imageat_edit_virtual-try-on, imageat_edit_city-teleport, imageat_edit_ai-edit-pro. New features appear automatically.

If the feature catalog can't be reached at startup, a single generic imageat_edit_image tool (taking a feature id parameter) is registered instead, so the server still works.

Setup

  1. Create an API key on your ImageAT Projects page (starts with iat_live_).
  2. Add the server to your MCP client config:
json
{  "mcpServers": {    "imageat": {      "command": "npx",      "args": ["-y", "@imageat/mcp"],      "env": {        "IMAGEAT_API_KEY": "iat_live_xxxxxxxxxxxx"      }    }  }}
  1. Restart the client. The imageat_* tools will be available.

Environment variables

VarRequiredDefaultNotes
IMAGEAT_API_KEYyes—Your iat_live_ key.
IMAGEAT_BASE_URLnohttps://imageat.comThe web app that serves the /api/v1/* generation endpoints. Point at http://localhost:3000 for local dev.

Remote server (claude.ai, ChatGPT, other web clients)

The npx setup above is stdio — it works in desktop apps that launch a local process (Claude Desktop, Cursor, Claude Code). Browser clients like claude.ai and chatgpt.com instead connect to a remote MCP endpoint over Streamable HTTP.

The same tools are served over HTTP by dist/http.js. Each MCP session carries the user's own API key via the Authorization: Bearer iat_live_... header (multi-user), so this is what you point a hosted connector at.

bash
# Run the remote server locally against a local ImageAT instance:IMAGEAT_BASE_URL=http://localhost:3000 PORT=8787 npm run start:http# MCP endpoint: http://localhost:8787/mcp   ·   health: /health

Deploy it (e.g. on Dokploy as mcp.imageat.com) with the included Dockerfile, then add it as a custom connector:

  • claude.ai — Settings → Connectors → Add custom connector → URL https://mcp.imageat.com/mcp.
  • ChatGPT — Connectors / Developer mode → add server URL https://mcp.imageat.com/mcp.

Browser clients run the OAuth flow: after adding the connector they open a consent page served by this server where you paste your iat_live_ key, and the client receives a short-lived access token bound to it. Non-browser clients (mcp-remote, MCP Inspector, curl) can still send a raw iat_live_ key directly as Authorization: Bearer iat_live_....

The OAuth layer (src/oauth.ts) is a minimal, stateless authorization server: it exposes the RFC 8414/9728 discovery docs, RFC 7591 dynamic client registration, and PKCE /authorize + /token. Access/refresh tokens are the API key encrypted (AES-256-GCM) under MCP_OAUTH_SECRET, so there is no session store.

VarWhereDefaultNotes
PORTremote only8787HTTP listen port.
IMAGEAT_BASE_URLbothhttps://imageat.comUpstream web app serving /api/v1/*.
MCP_OAUTH_SECRETremote only(random per boot)Long random string that encrypts issued OAuth tokens. Set this in production or tokens are invalidated on every restart.
MCP_PUBLIC_URLremote onlyderived from requestPublic origin, e.g. https://mcp.imageat.com. Only needed if proxy headers are wrong.

Local development

bash
npm installnpm run build
# Inspect the stdio server with the official MCP Inspector against a local ImageAT instance:IMAGEAT_API_KEY=iat_live_... IMAGEAT_BASE_URL=http://localhost:3000 \  npx @modelcontextprotocol/inspector node dist/index.js

Auth roadmap

Today the server authenticates with a static iat_live_ API key (the remote server reads it per-session from the Authorization header). Auth is isolated behind an AuthProvider interface (src/auth.ts), so a future "Sign in with ImageAT" OAuth provider can be dropped in without changing the tools, the HTTP client, or the backend /v1 routes.

來源：README.md，提交 a0adf12

工具

0
工具後設資料尚未被收錄。

版本歷史

1
  1. v0.4.0最新Sep 16, 2026