GenMagic

io.github.yumaheymans由 GitHub 帳號 yumaheymans 發布v1.1.1更新於 Oct 24, 2026

Generate images, video, speech, music and text with 450+ AI models, one API key and one balance.

發布方自己的服務：CareJobs.Now

已驗證Streamable HTTP可網頁執行媒體與設計AI 與機器學習

概覽

AI 產生的概覽

GenMagic 是託管式 MCP 伺服器，讓助理以 450+ 個 AI 模型生成文字、圖像、語音、音樂與影片，並從同一份預付餘額扣款。

可以做什麼
列出並篩選即時模型目錄，查看價格與能力；列出模型不需金鑰，生成時可指定模型 id。生成文字、程式碼、SVG 圖形或完整網頁。生成或編輯圖像，包含 logo 類品牌標誌。將文字逐字轉為語音，並依描述生成原創音樂。以提示詞或首幀圖像啟動非同步影片工作，再輪詢工作直到完成或失敗。每次結果都會回報所用模型、費用與剩餘額度。
適用情境
適合需要助理隨選產出媒體或文案的情境，例如圖像、配音、音樂、短影片與網頁，而不必逐一串接各模型供應商；也適合希望用單一預付餘額並逐次查看費用的使用者。
執行需求
遠端 Streamable HTTP 端點，無需安裝。需要 GenMagic 帳號（電子郵件已驗證）與 API 金鑰，以 Authorization Bearer 權杖送出；生成類工具必須提供金鑰，列出工具與模型目錄可匿名存取。需要能連線至該託管服務的網路。
安裝前請注意
生成會從預付美元餘額實際扣款（1 點 = 1 美分），影片在完成輪詢時計費。API 金鑰是有效憑證，須妥善保密。生成內容會送往第三方託管服務處理；開啟品牌個人化後，輸出會依所連結的網站或 LinkedIn 資料調整。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 GenMagic，將其新增到工作區。
  2. 連接 Authorization token 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/io.github.yumaheymans%2Fgenmagic/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": {
    "genmagic": {
      "type": "http",
      "url": "https://genmagic.co/api/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

GenMagic MCP Server

One Model Context Protocol server that lets any AI agent generate text, images, speech, music, and video, billed against a single prepaid credit balance.

GenMagic is one studio and one API for every generative AI model, in every medium. This repository is the connection guide and registry descriptor for its remote MCP server: point any MCP client at one URL, drop in your key, and your agent gets six generation tools plus list_models: it can pick any of GenMagic's 450+ models (GPT, Claude, Gemini, Veo, Kling, Seedance, FLUX, Recraft and more) by id, or leave the choice to GenMagic.

  • Endpoint: https://genmagic.co/api/mcp
  • Transport: Streamable HTTP (stateless JSON-RPC)
  • Version: 1.1.1
  • Auth: Authorization: Bearer gm_live_... (create a key at https://genmagic.co/get-api-key)
  • Hosted: nothing to install or run. It is a live remote server.
  • Registry: listed on the Official MCP Registry as io.github.yumaheymans/genmagic (see server.json).

Tools

ToolWhat it does
list_modelsThe live catalog, newest first: each model's id, price and key details (context window and capabilities for text, voices for speech, resolutions, clip lengths and sound for video). Filter by category, search or capability. No key needed.
generate_textGenerate text, code, an SVG graphic, or a full web page (optional type, optional model).
generate_imageGenerate or edit an image; type=logo for a brand mark; optional model.
generate_speechText to speech, voiced verbatim; optional model and voice.
generate_musicAn original music track from a description; optional model.
create_videoStart an async video job with any text-to-video model, duration, resolution and generate_audio it supports (returns a job id, charges nothing yet).
get_videoPoll a video job; charges once, on completion.

Every generate tool takes an optional model id from list_models. A model id that does not run that kind of output is refused with a pointer to list_models, never swapped for another model; leave model out and GenMagic picks a well-priced default.

initialize, ping, tools/list and list_models are open; the generate tools require your key. Every tool result self-describes its spend under the MCP _meta key genmagic.ai/usage (model that ran, cost_cents, credits_remaining, and a rate object), so an agent paces itself and tracks its budget from the result it already has.

Connect

Create a key at https://genmagic.co/get-api-key (sign up free and verify your email), then add the server to your client.

Claude Code

bash
claude mcp add --transport http genmagic https://genmagic.co/api/mcp \  --header "Authorization: Bearer gm_live_your_key"

VS Code

One-click install: open https://genmagic.co/developers#mcp and choose Add to VS Code; VS Code asks for your key once and stores it as a secret. Or add it to .vscode/mcp.json yourself:

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

Cursor, Windsurf, and most clients

Add a genmagic entry to your MCP config (.cursor/mcp.json, or the client's equivalent). Cursor also has a one-click Add to Cursor at https://genmagic.co/developers#mcp.

json
{  "mcpServers": {    "genmagic": {      "url": "https://genmagic.co/api/mcp",      "headers": { "Authorization": "Bearer gm_live_your_key" }    }  }}

Claude Desktop

Add GenMagic as a custom connector in Settings → Connectors, using the URL https://genmagic.co/api/mcp and a header Authorization: Bearer gm_live_your_key.

Any JSON-RPC client (raw)

The server is plain JSON-RPC over HTTP, so you can drive it with curl:

bash
# List the tools (no key required)curl https://genmagic.co/api/mcp \  -H "Content-Type: application/json" \  -H "Accept: application/json, text/event-stream" \  -d '{ "jsonrpc": "2.0", "id": 1, "method": "tools/list" }'
# Generate an image (key required)curl https://genmagic.co/api/mcp \  -H "Authorization: Bearer gm_live_your_key" \  -H "Content-Type: application/json" \  -H "Accept: application/json, text/event-stream" \  -d '{    "jsonrpc": "2.0", "id": 2, "method": "tools/call",    "params": {      "name": "generate_image",      "arguments": { "prompt": "a cobalt prism refracting into a spectrum, dark studio, 3D render", "size": "1024x1024" }    }  }'

How billing works

There is no separate agent meter. Every generation, from the studio, the REST API, or this MCP server, draws the underlying model's real provider cost from one prepaid USD balance (1 credit = 1 US cent). A new account gets a one-time $0.50 starter balance after email verification, one per person; top up in $5 to $100 increments, or subscribe for a renewing balance with a subscriber bonus and commercial-use rights. create_video and a still-rendering get_video poll charge nothing; the clip's cost lands on the poll that completes it. See https://genmagic.co/pricing.

On-brand generation

Turn on personalization once in the app (connect a website or LinkedIn) and every tool call comes out shaped by your brand's voice, audience, and colors, no prompt changes required. Visual tool results additionally report a brand_fit read-back so an agent can regenerate anything that drifted off-palette.

Links

License

MIT. See LICENSE. This repository is connection documentation and the registry descriptor; the GenMagic service itself is operated at genmagic.co.

來源：README.md，提交 a7c6f35

工具

7

list_models

List models

List the AI models GenMagic runs, newest first, with each one's id, price and key details (context window and capabilities for text, voices for speech, resolutions, clip lengths and sound for video). Pass a model's id as 'model' to generate_text, generate_image, generate_speech, generate_music or create_video. Filter by category and search; the model marked default is the one that runs when no model is named (for generate_text, when no type is given either). Needs no API key and costs nothing.

輸入參數limitsearchcategorycapability

generate_text

Generate text

Generate text with GenMagic: copy, an answer, a draft, code, an SVG graphic, or a complete web page. Pass 'type' to make a specific artifact ("website" for a full self-contained HTML page, "svg" for a vector graphic, "code" for a code snippet, "writing" for prose); omit it for plain text. Attach files for the model to read via 'attachments' (PDF, Word/Excel/PowerPoint, CSV, text/code, images, audio); GenMagic converts anything a model cannot ingest natively. If the caller's account has brand personalization on, the output comes out on-brand automatically, and a website or svg also picks up the brand's palette and typeface.

輸入參數typemodelprompt必填systemattachments

generate_image

Generate an image

Generate an image from a text description with GenMagic. Pass 'image' (an https or data URL) to transform THAT image instead of generating from scratch (image-to-image), and type "logo" to make a brand mark. Returns the image inline plus a hosted URL. On-brand automatically when the account has brand personalization on.

輸入參數sizetypeimagemodelprompt必填

generate_speech

Generate speech (text to speech)

Turn text into spoken audio with GenMagic. Returns the audio inline plus a hosted URL. The text is voiced verbatim.

輸入參數input必填modelvoice

generate_music

Generate music

Generate an original music track from a text description with GenMagic. Returns the audio inline plus a hosted URL. On-brand in mood automatically when the account has brand personalization on.

輸入參數modelprompt必填

create_video

Create a video

Start generating a video from a text description with GenMagic (or animate an image: pass 'image' as the opening frame), with any text-to-video model, clip length, resolution and sound the model supports (list_models with category "video" shows them and the per-second price). Video is asynchronous, so this returns a job id immediately (nothing is charged yet). Poll it with the get_video tool every few seconds until the status is "completed" to get the hosted video URL. A failed render is never charged. On-brand automatically when the account has brand personalization on.

輸入參數imagemodelprompt必填durationlast_frameresolutionaspect_ratiogenerate_audio

get_video

Check a video job

Check a video started with create_video. Returns "queued" or "processing" while it renders, "completed" with the hosted video URL when it is ready, or a terminal failure if the render failed. Poll every few seconds until it is either completed or failed (both are terminal: stop polling once you see one). The clip is charged once, on completion.

輸入參數id必填

版本歷史

3
  1. v1.1.1最新Oct 10, 2026
  2. v1.1.0Oct 9, 2026
  3. v1.0.0Sep 16, 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

其他 Carejobs 整合

CareJobs.Now

now.carejobs官方發布者

免費、唯讀的 MCP 伺服器，用來搜尋美國醫療照護產業的職缺與公告薪資。

Streamable HTTP可網頁執行
位置與生活6 個工具Oct 30, 2026

Bestrobotmower Mcp

由 GitHub 帳號 yumaheymans 發布

唯讀存取 BestRobotMower.co 的割草機器人資料集：規格、附日期的價格、0-5 評分、比較與依庭院條件的推薦。

Streamable HTTP可網頁執行
位置與生活4 個工具Nov 4, 2026