FlowMusicAI MCP Server

Flow Music AI - Free AI Music Generator for Full Songs

<p align="center"><a href="https://flowmusicai.app"><img src="./assets/hero.png" alt="FlowMusicAI" width="720" /></a></p>

A Model Context Protocol server that exposes the canonical FlowMusicAI knowledge surface — plans and pricing, pricing, docs, FAQ, official links — to MCP-compatible AI clients such as Claude Desktop, Cursor, Windsurf, and Continue. Read-only, no API keys, no quota, ~50 ms cold start.

Official website: https://flowmusicai.app

💳 About FlowMusicAI

Flow Music AI (flowmusicai.app) is a browser-based AI song generator that converts a plain-text description into a fully produced, mixed track in roughly 80 seconds. You type a one-line brief, optionally choose genre tags and a vocal mode, paste your own lyrics or let the built-in lyric assistant write them, then hit Generate. The result is a complete song — vocals, instruments, and arrangement — delivered as a downloadable MP3 with a royalty-free commercial license. No digital audio workstation, no music theory knowledge, and no post-production work is required.

Key Features

• Prompt-to-song in about 80 seconds. A single descriptive sentence drives the full composition. Vocals, instruments, and mixing are handled end-to-end by the model.
• 15+ stackable genre tags. Combine Pop with Lo-Fi, Cinematic with EDM, or any other pairing. A style-weight slider and a weirdness dial let you tune how strictly the model follows the tags.
• Lyrics assistant and custom lyric import. Describe a topic and the tool drafts lyrics for you, or paste your own text (up to 5,000 characters). Instrumental mode skips vocals entirely with one tap.
• Multiple model versions. The composer exposes V4 through V5.5 checkpoints, letting you choose between stability and the latest capabilities.
• Royalty-free commercial license with a PDF certificate. Every generated track is cleared for use in paid ads, YouTube uploads, Spotify distribution, and client deliverables. Paid plans add a downloadable certificate listing the prompt, model version, and generation timestamp.
• Instant MP3 download; WAV and stems on Pro. Free and entry plans deliver an unwatered MP3 the moment generation finishes. Pro and Studio tiers add WAV export and stem separation (vocals, drums, bass, mix bus). Failed runs refund credits automatically.
• C2PA-compatible inaudible watermark. Each track ships with a content-credential watermark that satisfies disclosure requirements from Spotify, YouTube, Apple Music, and TikTok without audible artifacts.

Use Cases

• Short-form video creators who need background tracks for TikTok, Reels, and YouTube Shorts that will not trigger content-ID claims or royalty splits.
• Podcasters building a custom intro or outro theme that matches their show's exact mood without a stock-library subscription.
• Indie game developers scoring menus, levels, boss fights, and trailers without hiring a composer or paying per-track license fees.
• Marketing teams and agencies running A/B tests across multiple audio beds for paid social, programmatic, and pre-roll campaigns on short turnaround.
• Filmmakers and video editors sourcing cinematic cues, montage beds, and trailer stingers at any mood without navigating synchronization licensing.
• Streamers and VTubers generating DMCA-clear stream loops, just-chatting background music, and intro stingers specific to their channel.
• Individuals creating personal gifts such as birthday songs, wedding tunes, or anniversary surprises built around custom lyrics.

Who Is It For

Flow Music AI targets people who need original, commercially cleared audio but have no music production background and no budget for session musicians or stock libraries. The core audience spans solo content creators, indie developers, small marketing teams, and freelance video producers. Because the workflow requires only a text description and takes under two minutes per track, it fits anyone on a tight deadline who cannot afford the time or cost of traditional music licensing. New accounts receive 65 free credits — enough to produce one complete song before any payment is required — and paid plans start at $9 per month, making it accessible to hobbyists and professionals alike.

Tools

get_pricing

Return the canonical pricing entry point for the site so an agent can cite plans without scraping. (FlowMusicAI)

Input: no parameters. Returns: text/markdown.

get_official_links

Return the canonical list of official links for FlowMusicAI (website, support, docs when available).

Input: no parameters. Returns: text/markdown.

Resources

• site://flowmusicai/plans — Public plan information surfaced on the pricing page.
• site://flowmusicai/pricing — Canonical pricing entry point.
• site://flowmusicai/docs — Canonical documentation entry point.
• site://flowmusicai/faq — Short FAQ generated from public site metadata.
• site://flowmusicai/links — Canonical URLs to share with users.

Prompts

tell_me_about_flowmusicai

Summarize what the site is, who it's for, and how it works. — FlowMusicAI

compare_pricing_flowmusicai

Compare available plans for a stated workload and pick one. — FlowMusicAI

Installation

Install via Smithery

npx -y @smithery/cli install flowmusicai-mcp --client claude

(Replace claude with cursor, windsurf, or continue for those clients.)

Install from source

git clone https://github.com/rocnubie/flowmusicai-mcp.git
cd flowmusicai-mcp
pnpm install

Then add to your MCP client config (claude_desktop_config.json for Claude Desktop, mcp.json for Cursor / Windsurf / Continue):

{
  "mcpServers": {
    "flowmusicai-mcp": {
      "command": "node",
      "args": [
        "/absolute/path/to/flowmusicai-mcp/src/index.mjs"
      ]
    }
  }
}

Debug with MCP Inspector

npx @modelcontextprotocol/inspector node src/index.mjs

Official Links

• Website: https://flowmusicai.app
• Pricing: https://flowmusicai.app/pricing
• Docs: https://flowmusicai.app/docs
• Community: https://discord.gg/flowmusicai
• GitHub: https://github.com/flowmusicai
• Support: support@flowmusicai.app

Development

pnpm install
pnpm start                 # run the server over stdio

License

MIT