More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Taoyuan Songwriter: an original song for one person

    1

    Tell your agent who the song is for; it writes the lyrics, shows you the plan, then delivers two finished versions with cover art via the Taoyuan MCP server.

    Free

    2 usesSecurity scanned
    Taoyuan Songwriter: an original song for one person

    Taoyuan Songwriter: an original song for one person

    Example session with this skill installed

    Write a birthday song for my mum, she's turning 60. She loves singing in the kitchen while she cooks. Lyrics in Chinese please.

    • Read your context and instructions
    • Compiled the taoyuan songwriter

    Summary

    I compiled a song plan from your two details (the kitchen, the singing), showed it to you, and after your go-ahead generated two finished versions. Total cost: 1 credit.

    The plan (shown before generating)

    Title把岁月煮香 (Simmering the Years)
    StyleMandopop ballad, 1980s Chinese folk-pop, clear female vocal, warm and celebratory, around 78 BPM, grand piano, fingerpicked nylon-string guitar, soft strings, brushed drums
    Lengthabout 2:30

    Lyrics

    [Verse | intimate, sparse]
    灶台亮起晨光
    锅盖跟着歌声响
    围裙兜住热饭香
    蒸汽绕过六十春
    
    [Chorus | bright, layered harmonies]
    妈祝你生日快乐
    妈祝你生日快乐
    你把岁月慢慢煮香
    一屋灯火都来尝
    
    [Final Chorus | bright, layered harmonies]
    妈祝你生日快乐
    灶火映红你的笑
    六十圈年轮正暖
    饭香还绕着屋梁
    
    [Outro | gentle, fading]
    妈祝你生日快乐
    歌声还在灶边亮
    

    Why this way: the verse opens on the stove at dawn because that is where she sings; the chorus turns cooking into the metaphor for sixty years ("you slowly simmered the years until they smelled sweet"). The 1980s folk-pop feel matches the music someone turning 60 grew up with.

    The result (after you said yes)

    Two complete versions, 2:29 each, with cover art:

    Version 1 (my pick, cleaner chorus harmony): https://media.musicsforyou.com/generations/0ca98933-349a-462d-ae27-af7c90c9e910/fdde1f37-508b-4799-b987-0ee6b958f88e.mp3

    Version 2 (warmer verse, slightly slower): https://media.musicsforyou.com/generations/0ca98933-349a-462d-ae27-af7c90c9e910/fdfdb8ab-8788-4bd9-b062-bdd87cce47d1.mp3

    Cover: https://media.musicsforyou.com/generations/0ca98933-349a-462d-ae27-af7c90c9e910/fdde1f37-508b-4799-b987-0ee6b958f88e.jpg

    Tool calls made: compile_music_prompt (free) → generate_music (1 credit) → get_generation with waitSeconds: 120 (free). Credits remaining: 2.

    Next steps

    • Want the final chorus bigger? I can redo just that section with replace_section (1 credit) and keep the rest.
    • Want a karaoke version for the party? separate_stems gives you vocals and instrumental separately (free).
    • Want an English verse for the grandchildren to sing? Say so and I'll revise the plan before generating again.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    What it does

    Taoyuan Songwriter teaches your agent to write an original song for one specific person or moment and turn it into finished audio. You say who the song is for and two or three true details. The agent compiles a plan (title, production-ready style description, full lyrics with section cues), shows it to you before anything is charged, revises on request, then renders two complete versions plus cover art in about 90 seconds.

    It connects to the Taoyuan MCP server (https://api.musicsforyou.com/mcp), the same engine behind musicsforyou.com, so the audio is real, downloadable MP3, not a prompt you have to paste somewhere else.

    Good for

    • Birthday, wedding, anniversary and farewell songs written around real memories
    • Podcast intros, jingles and theme music (instrumental or sung)
    • Turning a toast, a letter or a story into a song
    • Lyrics in English, Chinese, Japanese, Spanish and more; the agent matches the language to the person the song is for
    • Fixing one verse without redoing the whole song, and splitting vocals from the instrumental for karaoke

    How it works

    1. Brief: the agent asks at most one question if it does not know who the song is for. It never invents biographical facts.
    2. Plan (free): compile_music_prompt returns title, style and lyrics. You see all of it.
    3. Revise (free): "stronger chorus", "make it a duet", "shorter" - one instruction at a time, with locks for the parts you liked.
    4. Generate (1 credit): two candidates, each with audio and cover. The agent recommends one and says why.
    5. Follow-ups: redo one section (1 credit) or separate stems (free).

    What you need

    A Taoyuan account. Claude Code, Cursor, Codex CLI, claude.ai and ChatGPT sign in through the built-in OAuth flow; other clients use an API key from https://www.musicsforyou.com/developers. New keys include 3 free credits, enough for three songs. After that, credits are bought on musicsforyou.com; the agent hands you the purchase link and never buys on your behalf.

    What's inside

    • SKILL.md: the workflow, writing-quality rules and output format
    • references/connect.md: copy-paste setup for Claude Code, Cursor, Codex CLI, Claude Desktop (npx taoyuan-mcp), claude.ai and ChatGPT
    • references/tools.md: every tool, its cost and parameters
    • references/api.md: REST fallback for agents without MCP support
    • references/examples.md: two complete sessions, including the exact tool calls

    Honest limits

    • Songs come out between 1.5 and 5 minutes; a 20-second jingle means trimming the result.
    • Generation can fail on the provider side (rare). Failures are not charged and the agent retries once.
    • The skill does not write to your files or run shell commands. Its only network destination is api.musicsforyou.com.

    Made by the Taoyuan team (musicsforyou.com). Open-source MCP client: https://github.com/JerryYangAI/taoyuan-mcp

    Requires Taoyuan API key or MCP server access. Audio generation costs 1 credit per call. Rendering takes 60-120 seconds. Does not produce sheet music or MIDI files.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    Musicsforyou
    Taoyuan MCP
    Node.js

    Permissions

    Network Access

    Allowed Hosts

    api.musicsforyou.com

    Network access is only to api.musicsforyou.com (the Taoyuan MCP server / REST API) and www.musicsforyou.com (creating an API key, buying credits). Terminal is used only for the optional curl fallback in references/api.md when the agent has no MCP support. The skill never writes files and never reads environment variables other than TAOYUAN_API_KEY.

    Listed1 month ago
    Updated1 month ago

    What's inside

    Frequently Asked Questions

    This skill manages the entire production cycle, from structuring your personal stories into professional lyrics to generating two high-fidelity audio tracks and custom cover art, saving you from manual prompt engineering.

    Yes, this skill provides professional audio post-processing that separates your song into distinct vocal and instrumental stems, making it perfect for custom mixing or karaoke.

    The skill requires a connection to the Taoyuan (musicsforyou.com) API or MCP server; you will need an active API key and available credits to generate the final audio files.

    Absolutely—the skill includes a dedicated refinement stage where you can adjust the mood, vocal gender, or specific lyrics before the final audio generation to ensure the track meets your vision.

    The Taoyuan Songwriter is designed for compatibility with any AI agent platform that supports MCP servers or REST API integrations, including popular LLM orchestration frameworks.

    Each successful generation request produces two distinct full-length audio versions, a set of high-quality cover art, the final structured lyrics, and separated audio stems.

    New in Image & Video