minimax-music

MiniMax 音乐生成技能 - 支持歌词生成(Lyrics Generation)、音乐生成(Music Generation)。支持自定义歌词段落结构(verse/chorus/bridge/intro/outro)，异步生成 MP3 格式音乐，自动下载保存到本地。

Security Scan

The skill's description, SKILL.md, and code disagree about how it works and what credentials are required (missing metadata for MINIMAX_API_KEY, synchronous vs asynchronous behavior, and an unexpected query endpoint), so treat it as inconsistent until clarified.

Purpose & Capability

The skill claims to use the MiniMax API for music and lyrics generation (expected). However the registry metadata lists no required environment variables while both SKILL.md and the code clearly require MINIMAX_API_KEY (and optionally MINIMAX_REGION). That mismatch between claimed requirements and the actual code is inconsistent.

Instruction Scope

SKILL.md describes asynchronous generation returning task IDs and async workflows (generate_music -> task ids, query/wait), but the shipped Python implements generate_music as a synchronous POST that decodes audio and returns a local file path. The query_music_task implementation calls /query/video_generation (a video endpoint) which doesn't match the music endpoints described. These contradictions mean the runtime instructions and implementation do not align.

Install Mechanism

There is no install spec (instruction-only style plus a Python script), which is low risk. The script imports the third-party 'requests' library but the skill metadata does not declare dependencies—users must ensure requests is available in their environment.

Credentials

The code requires MINIMAX_API_KEY (and optionally MINIMAX_REGION), which is proportionate to the stated purpose. However the registry metadata failed to declare these required environment variables; that omission is a meaningful inconsistency and a usability/security concern (users might not realize they must provide an API key).

Persistence & Privilege

No elevated persistence requested: always is false, the skill doesn't request system-wide config changes, and it only reads/writes local files for downloaded audio (expected for this functionality).

What to consider before installing

Do not supply a real MINIMAX_API_KEY to this skill until the inconsistencies are resolved. Specific actions: (1) Ask the publisher to update registry metadata to declare MINIMAX_API_KEY and MINIMAX_REGION. (2) Request clarification/fix: does generate_music return a task ID (async) or an MP3 file (sync)? Fix the /query endpoint (video_generation looks wrong). (3) Inspect/run the Python script in an isolated environment (or sandbox) and ensure the 'requests' dependency is present. (4) If you must test, use a scoped or throwaway API key and run locally in a VM/container so the skill can't access other credentials or sensitive files. (5) Prefer installing only if the author corrects the documentation/metadata so behavior and required credentials match the code.

MiniMax Music Skill

使用 MiniMax API 进行音乐生成和歌词创作。支持歌词生成、音乐生成，支持自定义歌词段落结构，异步生成 MP3 格式音乐，可自动下载保存到本地。

环境配置

{
  "MINIMAX_API_KEY": "your-api-key",
  "MINIMAX_REGION": "cn" | "int"
}

• MINIMAX_API_KEY: MiniMax API 密钥
• MINIMAX_REGION: 区域设置，cn 为中国，int 为国际（默认 cn）

可用函数

generate_music(lyrics, description, model)

音乐生成

参数:

• lyrics: 歌词内容
• description: 音乐描述/风格
• model: 音乐模型（默认: music-01）

返回: 任务ID列表

示例: generate_music("[verse]\n蓝天白云下\n我们在奔跑\n[chorus]\n青春就是这样", "欢快的流行歌曲")

generate_lyrics(description, keywords, model)

歌词生成

参数:

• description: 歌曲描述/主题
• keywords: 关键词列表（可选）
• model: 歌词模型（默认: lyrics-01）

返回: 生成的歌词文本

示例: generate_lyrics("一首关于梦想的歌", ["坚持", "希望", "前行"])

query_music_task(task_id)

查询音乐生成任务状态

参数:

• task_id: 任务ID

返回: 任务状态信息 {status, music_url, ...}

状态值: Pending, Processing, Success, Fail

wait_for_music(task_id, poll_interval, max_wait)

等待音乐生成完成（轮询）

参数:

• task_id: 任务ID
• poll_interval: 轮询间隔秒数（默认: 10）
• max_wait: 最大等待时间（默认: 600）

返回: 最终任务状态信息

download_music(music_url, output_path)

下载音乐到本地

参数:

• music_url: 音乐下载URL
• output_path: 保存路径

返回: 保存的文件路径

使用示例

直接生成音乐

python scripts/music.py generate "[verse]\n月亮挂在天上\n星星闪闪发光\n[chorus]\n夜空中最亮的星" -d "温柔的民谣歌曲"

先生成歌词，再生成音乐

# 先生成歌词
python scripts/music.py lyrics "一首关于夏天的歌" -k 阳光 沙滩 海浪

# 使用生成的歌词创建音乐
python scripts/music.py generate "<生成的歌词>" -d "欢快的夏季主题歌曲"

查询任务状态

python scripts/music.py query <task_id>

等待音乐完成

python scripts/music.py wait <task_id> -i 10

下载音乐

python scripts/music.py download <music_url> -o song.mp3

歌词格式说明

歌词按段落返回，使用空行分隔不同段落。可用的段落标记：

• [verse] - 主歌
• [chorus] - 副歌/高潮
• [bridge] - 桥段
• [intro] - 前奏
• [outro] - 尾奏

注意事项

1. 歌词越详细，生成的音乐效果越好
2. 可以不提供歌词，仅通过 description 生成音乐
3. 音乐生成是异步的，需要通过 task_id 查询状态
4. 建议使用 wait_for_music 函数自动等待完成
5. 音乐生成可能需要几分钟时间，请耐心等待