Skip to content

Make images and video in Claude with MagicShot

Add MagicShot to Claude once, then just ask. No key to paste: you sign in to MagicShot and it runs on your credits.

Paste this into Claude Code with your API key, then just ask.

terminal
claude mcp add --transport http magicshot https://mcp.magicshot.ai/mcp \
  --header "Authorization: Bearer msk_YOUR_KEY"

Run it once in your terminal. That is the whole setup.

Paste this into Cursor with your API key, then just ask.

~/.cursor/mcp.json
{
  "mcpServers": {
    "magicshot": {
      "url": "https://mcp.magicshot.ai/mcp",
      "headers": { "Authorization": "Bearer msk_YOUR_KEY" }
    }
  }
}

Add to your global mcp.json, or a per-project .cursor/mcp.json.

Paste this into VS Code with your API key, then just ask.

.vscode/mcp.json
{
  "servers": {
    "magicshot": {
      "type": "http",
      "url": "https://mcp.magicshot.ai/mcp",
      "headers": { "Authorization": "Bearer msk_YOUR_KEY" }
    }
  }
}

Note the servers key and "type": "http".

Paste this into Codex with your API key, then just ask.

~/.codex/config.toml
[mcp_servers.magicshot]
url = "https://mcp.magicshot.ai/mcp"
http_headers = { "Authorization" = "Bearer msk_YOUR_KEY" }

Streamable HTTP. On older Codex versions, add experimental_use_rmcp_client = true at the top of the file.

Paste this into Windsurf with your API key, then just ask.

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "magicshot": {
      "serverUrl": "https://mcp.magicshot.ai/mcp",
      "headers": { "Authorization": "Bearer msk_YOUR_KEY" }
    }
  }
}

Same idea as Cursor, but the key is called serverUrl.

Paste this into Claude Desktop with your API key, then just ask.

claude_desktop_config.json
{
  "mcpServers": {
    "magicshot": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.magicshot.ai/mcp",
               "--header", "Authorization: Bearer msk_YOUR_KEY"]
    }
  }
}

Connected MagicShot on claude.ai already? It shows up in the desktop app too, no config needed. Otherwise bridge it with mcp-remote so your key is passed through.

Paste this into your MCP client with your API key, then just ask.

the essentials
# transport  Streamable HTTP (POST + SSE)
# url        https://mcp.magicshot.ai/mcp
# auth       Authorization: Bearer msk_YOUR_KEY
# tools      suggest, search_tools, list_models,
#            text_to_image, text_to_video, image_to_video,
#            run_tool, get_job_status, get_credits

Anything that speaks MCP and lets you set a URL and an Authorization header.

https://mcp.magicshot.ai/mcp

How to connect

Connect MagicShot to Claude in under two minutes.

  1. Sign in to Claude

    Open claude.ai in your browser and sign in to your account.

    Open claude.ai
  2. Add MagicShot as a connector

    Press Connect MagicShot, or go to Customize, Connectors, Add custom connector, name it MagicShot and paste the endpoint. Then sign in to MagicShot when Claude asks.

    Add MagicShot to Claude
  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into Claude Code

    Copy the command above, swap msk_YOUR_KEY for your key, and run it.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into Cursor

    Copy the config above into ~/.cursor/mcp.json and swap msk_YOUR_KEY for your key.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into VS Code

    Copy the config above into .vscode/mcp.json and swap msk_YOUR_KEY for your key.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into Codex

    Copy the config above into ~/.codex/config.toml and swap msk_YOUR_KEY for your key.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into Windsurf

    Copy the config above into ~/.codeium/windsurf/mcp_config.json and swap msk_YOUR_KEY for your key.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Paste it into Claude Desktop

    Copy the config above into claude_desktop_config.json and swap msk_YOUR_KEY for your key.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

  1. Get your API key

    Sign in to MagicShot and create a key under API Keys. It looks like msk_ followed by letters and numbers.

    Create a key
  2. Add the server to your client

    Point it at the endpoint over Streamable HTTP and send your key as Authorization: Bearer msk_YOUR_KEY.

  3. Ask for an image or a video

    Describe what you want in plain words. Your assistant picks the tool, runs it on your credits and hands back the result.

MagicShot

What your assistant can do once connected

Nine tools, one connection. Your assistant chooses between them; you just say what you want.

  • Generate images

    Turn a prompt into finished images.

    text_to_image
  • Generate videos

    Turn a description into a clip.

    text_to_video
  • Animate a still

    Bring a photo to life with motion.

    image_to_video
  • Run any tool

    Background removal, upscaling, headshots and the rest of the catalog.

    run_tool
  • Find the right tool

    Search every MagicShot tool by what you need.

    search_tools
  • Get a recommendation

    The best tool and model for your goal.

    suggest
  • Browse the models

    Every model, with its ratios, durations and inputs.

    list_models
  • Track a job

    Follow a generation to the finish.

    get_job_status
  • Check your balance

    See the credits you have left, anytime.

    get_credits

Things you can just ask for

Pick a prompt to preview what your assistant sends back.

Your AI assistant

A cinematic sunset over a mountain lake, 16:9

Using MagicShot · text_to_image

The result comes back as a link, ready to download or drop into your project.

The latest models, all through one endpoint

Wan 3.0 Prime

Featured

Alibaba · Video

Wan 3 Prime is the fast tier of Alibaba's Wan 3.0 video model, built for text-to-video and image-to-video generation. It turns a prompt or a single photo into a video clip up to 30 seconds long, complete with a matching audio track.

Gemini Omni 1.1

Featured

Google · Video

Gemini Omni 1.1 is Google's fast multimodal video model, built to turn a text prompt or a single photo into a moving clip with sound already baked in. Inside MagicShot, it runs as the speed option for text-to-video and image-to-video, so you get a usable draft in seconds instead of minutes.

Qwen Image 3

Featured

alibaba · Image

Qwen Image 3 is Alibaba's third-generation image model, built to turn a single detailed prompt into dense, information-heavy visuals like newspaper pages, infographics, and exam papers. Inside MagicShot, it handles the kind of layout-heavy, text-heavy images that most image generators fall apart on.

MiniMax H3

Featured

MiniMax · Video

MiniMax H3 is an omni-modal video model built by MiniMax that reads text, images, video, and audio together and turns them into finished video clips. On MagicShot, it works as a text-to-video and image-to-video engine that renders motion, on-screen text, and native stereo sound in a single generation.

Wan 3.0

Featured

alibaba · Video

Wan 3.0 is Alibaba Tongyi Lab's text-to-video model, built to turn a written prompt, a photo, or a reference clip into a finished video with sound already in it. Inside MagicShot, it runs on paid plans for product demos, social clips, and short marketing videos.

Seedance 2.5

Featured

Bytedance · Video

Seedance 2.5 is ByteDance's flagship multimodal video model, available inside MagicShot on a paid plan. It turns a text prompt plus up to 50 image, video and audio references into a single clip of up to 30 seconds, with dialogue, music and sound effects generated alongside the picture.

What is MCP, without the jargon

MCP is a universal adapter for AI assistants. Plug MagicShot in once and your assistant gains a new skill: making images and video, and running MagicShot tools on your pictures, right inside the chat.

  • No code

    One click in Claude, or a one-time paste in your editor.

  • Same credits

    Runs on the MagicShot plan you already have.

  • Plain words

    Ask the way you would ask a person. It handles the rest.

MCP questions, answered

MCP (Model Context Protocol) is an open standard that lets an AI assistant use outside tools. Connecting MagicShot gives your assistant new abilities: making images and video, running MagicShot tools on your pictures, following a job and reading your credit balance. You keep chatting like normal; it just can do more.

Press Connect MagicShot on this page while signed in to claude.ai. Claude opens its add connector screen with MagicShot filled in; confirm it and sign in to MagicShot. You can also add it by hand under Customize, Connectors, Add custom connector, using the endpoint shown above.

Claude on the web and desktop through the connector, plus any MCP client that supports remote servers with a custom header: Claude Code, Codex, Cursor, Windsurf and VS Code among them.

For Claude, no: you sign in to MagicShot when you add the connector. For editors and command line tools, yes: create one in your MagicShot dashboard under API Keys and paste it into the config as a Bearer token. Keep it secret, since anyone holding the key can spend your credits. You can revoke and rotate keys anytime.

Generations made over MCP spend the same credit pool as the MagicShot app, on the plan you already have. There is no separate charge for using the server.

You do not have to choose. Describe your goal and your assistant calls suggest or search_tools to find the right one, then list_models for its options.

Generate without leaving your assistant

Connect MagicShot to Claude in one click, or paste one config into your editor.