Repository avatar
AI Tools
v1.6.0
active

video-generator

ai.makeaivideo/video-generator

AI video generator: make TikTok, Reels and Shorts videos with voiceover and captions, and post them.

Documentation

MakeAIVideo MCP Server: AI video generation for ChatGPT, Claude, Cursor and any AI agent

npm license MCP Registry

Add to Cursor Install in VS Code

The MakeAIVideo MCP server lets ChatGPT, Claude, Cursor and any AI agent create short-form AI videos (script, AI voiceover, AI or stock scenes, captions and music) from a prompt or a script, return the finished MP4, and post or schedule it to TikTok, Instagram Reels, YouTube Shorts and other connected social accounts.

It is the official Model Context Protocol server for MakeAIVideo, an AI video generator for short-form video.

Hosted serverhttps://mcp.makeaivideo.ai (Streamable HTTP, OAuth sign-in, no key to copy)
Local servernpx -y @makeaivideo/mcp (stdio, needs MAKEAIVIDEO_API_KEY)
Tools70: create, plan, edit, deliver and post videos
Works withChatGPT, Claude, Claude Code, Cursor, VS Code, Codex, Gemini CLI, Windsurf, Grok, Perplexity, Le Chat, Zed, Cline and any MCP client
OutputMP4 in 9:16, 1:1 or 16:9, with captions, in 24 languages
AccountA MakeAIVideo account with credits (pricing)
Registry nameai.makeaivideo/video-generator

What you can make

Ask your assistant for a video in plain words and it picks the right tool:

Image to video and character swap need a file upload, so they run in the web app and the REST API, not through MCP tools. See more on the examples page.

Example prompts

  • "Make a 30-second vertical explainer video about why octopuses have three hearts and give me the MP4."
  • "Give me 10 video ideas for a personal finance TikTok channel, then make the best one."
  • "Write a script for a UGC-style ad for my skincare serum. Show me the script before you make the video."
  • "Turn the blog post at this link into a 60-second YouTube Short." (then paste the URL)
  • "Make a 20-second spokesperson video announcing our new opening hours, with a calm British voice."
  • "How many credits would a 45-second list video cost?"
  • "When the video is ready, post it to my TikTok and YouTube tomorrow at 9am with a short caption."

Quick start

The hosted server is the easiest way in. Add https://mcp.makeaivideo.ai, sign in to MakeAIVideo in the browser window that opens, and press Approve. You do not paste a key. The steps for every app, with copy buttons, are on the AI Assistants page in the app.

After you connect, try: "How many MakeAIVideo credits do I have? Then give me three video ideas for my channel."

Claude (claude.ai and Claude Desktop)

  1. In Claude, open Customize → Connectors, press +, then Add custom connector.
  2. Name it MakeAIVideo and paste https://mcp.makeaivideo.ai.
  3. Press Add, then Connect. Sign in to MakeAIVideo and press Approve.

It then works in the Claude phone apps too. On Team and Enterprise plans an Owner adds connectors first, in Organization settings.

ChatGPT

  1. On the web, switch on Developer mode: Settings → Apps → Advanced settings. Custom connectors stay hidden until it is on.
  2. In Settings → Apps, choose Create. Name it MakeAIVideo and paste https://mcp.makeaivideo.ai.
  3. Set authentication to OAuth, press Create, then sign in to MakeAIVideo and approve.

Developer mode needs a Plus, Pro, Business, Enterprise or Education plan.

Claude Code

claude mcp add --scope user --transport http makeaivideo https://mcp.makeaivideo.ai

Then type /mcp in Claude Code, choose Authenticate, and sign in. Keep --scope user so the server works in every folder.

Cursor

Click Add to Cursor above and confirm Install, or add this to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "makeaivideo": { "url": "https://mcp.makeaivideo.ai" }
  }
}

Approve the sign-in window Cursor opens.

VS Code (GitHub Copilot agent mode)

Click Install in VS Code above, or run MCP: Add Server from the Command Palette, choose HTTP and paste the address. Or add this to .vscode/mcp.json:

{
  "servers": {
    "makeaivideo": { "type": "http", "url": "https://mcp.makeaivideo.ai" }
  }
}

Sign in when VS Code asks, then open Chat in Agent mode.

Codex

codex mcp add makeaivideo --url https://mcp.makeaivideo.ai
codex mcp login makeaivideo

Editing ~/.codex/config.toml by hand? The header is [mcp_servers.makeaivideo], with an underscore.

Gemini CLI

gemini mcp add --scope user --transport http makeaivideo https://mcp.makeaivideo.ai

Start Gemini CLI, run /mcp auth makeaivideo, then sign in. Or install the MakeAIVideo extension from this repo:

gemini extensions install https://github.com/makeaivideo-ai/mcp

Windsurf

Add this to ~/.codeium/windsurf/mcp_config.json, then fully quit and reopen Windsurf and approve the sign-in:

{
  "mcpServers": {
    "makeaivideo": { "serverUrl": "https://mcp.makeaivideo.ai" }
  }
}

Other apps

AppHow to connect
Grokgrok.com → Connectors → New Connector → Custom, paste the address
PerplexitySettings → Connectors → custom connector, OAuth and Streamable HTTP (Pro, Max or Enterprise)
Le ChatConnectors → + Add Connector → Custom MCP Connector
GitHub Copilot CLIcopilot mcp add --transport http makeaivideo https://mcp.makeaivideo.ai
Zed"context_servers": { "makeaivideo": { "url": "https://mcp.makeaivideo.ai" } } in ~/.config/zed/settings.json
ClineUses an API key: "type": "streamableHttp", the address, and the header Authorization: Bearer mav_...
n8n, Zapier, MakeAdd an MCP Client with the address (n8n works best with a Bearer API key)
All coding agents at oncenpx add-mcp https://mcp.makeaivideo.ai

Agents that sign themselves in can follow auth.md (OAuth with PKCE, or the device code flow). Prefer an API key? Send Authorization: Bearer mav_... and create the key in the app under Developers.

Local server (npx)

For apps that only run local commands. The package is a small stdio bridge to the hosted server, so the tools are always the same as the hosted ones. Node 18 or later.

{
  "mcpServers": {
    "makeaivideo": {
      "command": "npx",
      "args": ["-y", "@makeaivideo/mcp"],
      "env": { "MAKEAIVIDEO_API_KEY": "mav_YOUR_KEY" }
    }
  }
}

Works with Claude Desktop, Cursor, Windsurf, Cline, Zed, Continue and any stdio MCP client.

With Docker (the Dockerfile in this repo):

docker build -t makeaivideo-mcp .
docker run -i --rm -e MAKEAIVIDEO_API_KEY=mav_YOUR_KEY makeaivideo-mcp

Without a key the local server still starts and lists its tools from the bundled tools.json, so clients can inspect it. Every tool call then asks for a key.

VariableDefaultPurpose
MAKEAIVIDEO_API_KEY(required for tool calls)Your mav_ API key
MAKEAIVIDEO_MCP_URLhttps://mcp.makeaivideo.aiOverride the remote endpoint

Tools (70)

Create videos

ToolWhat it does
make_videoThe one-call path: give a topic, it prices the video and starts it when the balance covers it
create_explainer_videoExplainer video: one idea, explained clearly from question to payoff
create_listicle_videoList video: numbered tips or ranked picks, one scene per item
create_story_videoStory video for history, mysteries and true events
create_ugc_videoUGC-style ad that reads like a real recommendation
create_demo_videoProduct demo that shows the job getting done, step by step
create_article_videoTurns a blog post, newsletter or other writing into a short
create_spokesperson_videoOne lip-synced presenter, straight to camera, from a character you pick
create_video_from_scriptRenders a video from a script you already have
image_to_videoBrings one photo to life as a 5, 10 or 15 second clip; add narration and a face in the photo speaks it
talking_avatarThe person in your own filmed video says a new script (voice and lip sync onto your clip)
character_swapThe person in your 3-60 second video becomes a character or the person in your photos; motion and audio kept
upload_image_from_urlImports an image from a public URL (free) for image_to_video, character_swap or product photos
upload_video_from_urlImports a video from a public URL (free) for talking_avatar or character_swap

Each create tool takes the length, aspect ratio (9:16, 1:1 or 16:9), and an optional character and voice. It returns a video id at once; the render takes minutes.

Plan and write

ToolWhat it does
estimate_video_costQuotes the credit cost before anything is spent (no credits used)
write_scriptWrites the script only: title, hook and scenes with narration and visuals
structure_textTurns prose or a hand-written script into a scene script
fetch_articleReads a public article's text so it can become a video
generate_ideasScored video ideas for a niche and audience
enhance_promptRewrites a rough topic into a stronger brief

Track and deliver

ToolWhat it does
wait_for_videoWaits up to about 50 seconds per call and returns done, progress and the download link
get_videoEverything about one video: status, the full script with every scene (narration, picture, thumbnail, clip, lip-synced or not), music, voice and any edit in progress
get_video_statusProgress of one video, and its download link when ready
list_my_videosYour videos with id, title and status
get_download_urlA signed MP4 link for a finished video
get_captionsThe video's captions as SRT or VTT
create_share_linkA public link to the MP4 that expires (24 hours by default, up to 7 days)
export_videoRenders the video again after edits

Edit and redo

ToolWhat it does
update_videoEdits the title, topic, script or caption settings
generate_videoStarts or re-runs generation for a draft
regenerate_sceneRedoes one scene, optionally with new narration
edit_scene_narrationChanges what one scene says: the line is re-recorded and the picture is kept (cheaper than a remake)
edit_scene_visualChanges what one scene shows: the scene is remade from a new picture description, the line kept
set_musicChanges or removes a video's background music, or its volume (free; then export_video)
report_videoReports a bad video to our team for review (free)
regenerate_voiceRe-records the voiceover with another voice
cancel_videoStops a render and releases its reserved credits
delete_videoDeletes a video and its files

Characters, voices and music

ToolWhat it does
list_charactersPreset and saved characters (presenters)
generate_character_portraitsPortrait options for a new character from a description
generate_voice_previewsVoice options for a new character
create_characterSaves a character from a chosen portrait and voice
list_voicesVoices, filtered by gender, accent, age and use case
preview_voiceHear a voice read your text
generate_musicA background track (15, 30 or 60 seconds) from a prompt and mood
list_musicYour generated tracks
list_music_libraryThe background music library, filtered by mood, with previews (free)
update_characterRenames one of your characters, changes its description or its voice
delete_characterDeletes one of your own characters
list_templatesThe template library

Post to social media

ToolWhat it does
list_social_accountsYour connected social accounts and their ids
get_connect_linkA link a person opens in the browser to connect an account
publish_videoPosts a finished video now or at a scheduled time, with one caption or one per platform
get_publish_statusResult of a post, with the live post URLs
list_postsPast and scheduled posts
suggest_captionA ready-to-post caption with hashtags, written from the video's script (free)
reschedule_postChanges when a scheduled post goes out
cancel_postCancels a scheduled post on every platform

Account and automation

ToolWhat it does
whoamiWho is signed in, the plan and the credit balance
get_creditsThe credit balance in detail, when the plan credits refill, and the add-credits link
get_top_up_linkA link you open in a browser to add credits (you pay on Stripe's page; the assistant cannot pay)
create_webhookCalls your URL when a video is ready, fails, or is posted, or when credits run low or refill
list_webhooksYour webhooks and their delivery health
test_webhookSends one signed test event to a webhook's URL
list_webhook_deliveriesA webhook's last 50 delivery attempts
delete_webhookDeletes a webhook
get_workspaceThe workspace's name and saved defaults
update_workspaceRenames the workspace or changes its saved defaults
get_brand_kitThe workspace watermark and whether new exports carry it
set_watermarkSets, shows, hides or removes the logo watermark from a public image URL

How a video gets made

  1. Quickest: the assistant calls make_video with your topic. It prices the video and starts it when your balance covers it. For more control it quotes the cost with estimate_video_cost, then calls a create_*_video tool with your brief, or write_script first so you can edit the script.
  2. If a create result says started: false, nothing is rendering or charged: held says why, with priced options and the next call to make. Out of credits, it names the shortfall and when the plan refills, and get_top_up_link gives you a link to add more.
  3. It calls wait_for_video (each call waits up to about 50 seconds) until done is true, usually a few minutes.
  4. It returns the MP4 link from get_download_url, or posts the video with publish_video.
  5. To fix a finished video, get_video shows every scene; edit_scene_narration, edit_scene_visual, regenerate_scene and set_music change one part, and export_video (or auto_export) updates the file.

Posting to TikTok, Instagram, YouTube and more

publish_video posts a finished video to one or more connected accounts, now or at a scheduled time. Supported: TikTok, Instagram (as a Reel), YouTube (a 9:16 video posts as a Short, 16:9 as a regular video), Facebook (Pages only), LinkedIn, X (videos up to 140 seconds), Threads, Pinterest, Bluesky, Telegram and Discord.

A person connects each account once. get_connect_link returns a short-lived link that you open in a browser to approve access; the assistant cannot do this step for you. Bluesky, Telegram and Discord connect in the web app. Posts are public, so a well-behaved assistant confirms the accounts and caption with you first.

Credits and pricing

The MCP server has no separate charge. It needs a MakeAIVideo account, and videos are paid from that account's credits, the same balance the web app uses.

  • New accounts start with a 7-day trial on a monthly plan. A card is required, and the trial includes trial credits.
  • estimate_video_cost quotes the exact cost before anything is spent. Checking status, listing and quoting use no credits.
  • An agent cannot create an account for you. Sign up at app.makeaivideo.ai/signup, then connect.

Plans and credit costs: makeaivideo.ai/pricing.

FAQ

Is there an MCP server for AI video generation?

Yes. The MakeAIVideo MCP server at https://mcp.makeaivideo.ai gives any MCP client 70 tools to write scripts, create finished short-form AI videos with voiceover, scenes, captions and music, and post them to social media. It is listed in the official MCP Registry as ai.makeaivideo/video-generator.

How do I make AI videos with ChatGPT?

Add https://mcp.makeaivideo.ai as a custom connector in ChatGPT (Developer mode, Settings → Apps → Create, authentication OAuth), sign in to MakeAIVideo, then ask ChatGPT for the video you want, for example "make a 30-second explainer about black holes". ChatGPT creates it and gives you the MP4 link. More on the ChatGPT video generator page.

Can Claude create TikTok videos?

Yes. Connect MakeAIVideo in Claude (Customize → Connectors → Add custom connector) and ask for a vertical TikTok video. Claude writes the script, makes the video with voiceover, captions and music, and can post it to your TikTok account now or on a schedule once you have connected it.

Can Cursor, Claude Code or other coding agents make videos?

Yes. Use the one-click Cursor or VS Code buttons above, or the one-line commands for Claude Code, Codex and Gemini CLI. For shell-only agents there is also the MakeAIVideo CLI.

Can an AI agent post videos to TikTok, Instagram Reels and YouTube Shorts?

Yes. After you connect each account once in the browser, the assistant can post or schedule a finished video to TikTok, Instagram, YouTube, Facebook Pages, LinkedIn, X, Threads, Pinterest, Bluesky, Telegram and Discord.

Can it make talking avatar videos?

Yes. create_spokesperson_video makes a lip-synced presenter video from a character you pick with list_characters or create with create_character. See talking avatar.

Can it turn an image into a video, or swap the person in my video?

Yes. Give the assistant a public link to the photo or video: upload_image_from_url or upload_video_from_url imports it, then image_to_video, talking_avatar or character_swap makes the video. See image to video and character swap.

Can it fix one scene without making the whole video again?

Yes. get_video lists every scene. edit_scene_narration changes a scene's line and keeps the picture, edit_scene_visual or regenerate_scene remakes the picture, and set_music swaps the music. Paid edits show the price first and charge nothing until the assistant confirms it with max_credits.

Do I need an API key?

Not for the hosted server: most apps sign you in with OAuth. You need a mav_ API key only for the local npx server, Cline, unattended automations, or apps without OAuth.

How much does it cost?

MakeAIVideo is a paid product. New accounts start with a 7-day trial (card required), and videos use credits from your plan. The MCP server itself has no extra charge. estimate_video_cost quotes each video first. See pricing.

How long does a video take, and what formats do I get?

A few minutes. You get an MP4 in 9:16 (vertical), 1:1 or 16:9, with captions burned in, and SRT or VTT caption files on request. Scripts, voiceover and captions can be in 24 languages.

Is it open source?

This package, the local stdio bridge, is MIT-licensed. The video generation runs on the hosted MakeAIVideo service.

Links

Related packages

PackageWhat it is
@makeaivideo/mcpThis MCP server, for ChatGPT, Claude, Cursor and other AI assistants
@makeaivideo/cliCommand line tool for terminals and shell agents: brief in, MP4 out
@makeaivideo/sdkTypeScript and JavaScript SDK for the REST API

License

MIT © MakeAIVideo

NPM
@makeaivideo/mcp
Install Command
npm install @makeaivideo/mcp

Remote