一个用于通过 Suno 生成音乐的 MCP (Model Context Protocol) 服务器。
生成歌曲：支持自定义歌词和风格标签
根据描述生成：让 Suno AI 编写歌词
等待完成：获取带有 Markdown 嵌入的音频 URL
下载 MP3：下载到本地文件夹
自动刷新令牌：令牌过期时自动刷新
Related MCP server: Suno-MCP
Node.js 18+
BrowserOS（或任何带有 --remote-debugging-port 的 Chromium），运行在 http://localhost:9100 且已登录 Suno。可通过 CDP_URL 覆盖。
Suno 账户（免费或付费均可）
npm install添加到你的 .mcp.json 中：
{
"suno": {
"command": "node",
"args": ["/path/to/suno-mcp/src/index.js"]
}
}工具 | 描述 |
| 使用自定义歌词和风格生成 |
| AI 根据描述编写歌词 |
| 等待生成完成，返回 URL + Markdown |
| 检查生成状态 |
| 获取账户积分信息 |
| 列出最近的歌曲 |
| 将 MP3 下载到文件夹 |
Suno 使用 Clerk 进行身份验证，并使用短效 JWT 令牌。客户端通过 CDP 在活动的 Suno 标签页中调用 window.Clerk.session.getToken() —— Clerk 会在后台处理刷新，因此无需重新加载页面。令牌会在其 exp 声明过期前 5 分钟进行缓存。
由于 hCaptcha 的存在，直接调用 API 会失败。客户端不是抓取表单选择器，而是驱动 Suno 自己的 React 应用：
通过 CDP 连接到 Suno 标签页；如有必要，导航至 /create。
从“创建”按钮遍历 React fiber 树，找到将 onCreateClick / lyrics / styles / mode 作为 props 暴露的组件 —— 即语义化的生成处理程序。
使用与 React 兼容的原生 setter 填充歌词和风格文本区域，从而更新内部状态。
等待 fiber 的 props 反映出新值（即 React 已重新渲染）。
通过 CDP Input.dispatchMouseEvent 在按钮坐标处分发受信任的鼠标点击事件。hCaptcha 将 CDP 合成的事件视为真实的用户手势（直接从 JS 调用 onCreateClick() 则不行 —— 验证码会拒绝）。
通过 CDP Network 事件捕获 /api/generate/v2 的响应，以返回新的片段 ID。如果错过了网络事件，则回退到轮询 /api/project/default。
这种方法对于 UI 改版具有鲁棒性：它依赖于语义化的 React props (onCreateClick, lyrics, styles) 和 aria-label="Create song" 按钮 —— 这两者都是核心功能且不太可能改变 —— 而不是依赖类名、textContent 匹配或 DOM 布局。
// Generate a song
const result = await suno_generate_song({
lyrics: "[Verse]\nHello world\n\n[Chorus]\nThis is a test",
style: "indie folk, acoustic, male vocals"
});
// Wait for it to complete
const songs = await suno_wait_for_songs({
ids: result.songIds
});
// Download it
await suno_download_song({
songId: songs[0].id,
folder: "./downloads"
});标题字段无法可靠工作（Suno UI 限制 —— 目前未通过新的生成路径连接）。
generate_from_description（简单模式）如果找不到 UI 模式切换开关，目前会回退到自定义模式；为了获得确定性的行为，请通过 generate_song 传递歌词和风格。
需要 BrowserOS 且保持活跃、可见的 Suno 会话。创建按钮在屏幕上必须具有非零尺寸，以便受信任的点击能够生效。
令牌每约 1 小时过期一次（通过 Clerk 自动刷新）。
MIT
Check the generation status of songs by their IDs
| Name | Required | Description | Default |
|---|---|---|---|
| ids | Yes | Array of song IDs to check |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It states the action ('Check status') but doesn't add context such as rate limits, authentication needs, expected response format, or whether it's a read-only operation. This is a significant gap for a tool with zero annotation coverage.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence that directly states the tool's purpose without any wasted words. It's appropriately sized and front-loaded, making it easy for an agent to parse quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of checking song generation status, the description is incomplete. With no annotations and no output schema, it fails to explain behavioral traits like response format or error handling. It relies solely on the input schema, leaving gaps in understanding how to interpret results or handle failures.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, with the 'ids' parameter fully documented in the schema. The description adds no additional meaning beyond what the schema provides (e.g., it doesn't specify ID format or constraints), so it meets the baseline of 3 where the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('Check') and resource ('generation status of songs'), making the purpose specific and understandable. However, it doesn't explicitly differentiate from sibling tools like 'suno_wait_for_songs' or 'suno_get_recent', which might also involve status checking or retrieval operations, so it doesn't reach the highest score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention prerequisites, timing (e.g., after generation), or compare to siblings like 'suno_wait_for_songs' for waiting or 'suno_get_recent' for recent songs, leaving the agent with no usage context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Download a song MP3 to a specified folder
| Name | Required | Description | Default |
|---|---|---|---|
| songId | Yes | The song ID to download | |
| folder | Yes | The folder path to save the MP3 to | |
| filename | No | Optional filename (without .mp3 extension). Defaults to song title or ID. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It states the tool downloads an MP3 file, implying a read/write operation that saves data locally, but doesn't cover critical aspects like permissions needed, file overwriting behavior, error handling (e.g., if the songId is invalid), or network/rate limits. This leaves significant gaps for a tool that modifies the filesystem.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence that front-loads the core action ('Download a song MP3') and specifies the destination. There is zero waste, and every word earns its place, making it highly concise and well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (a download operation that writes files) and lack of annotations and output schema, the description is incomplete. It doesn't explain what happens on success (e.g., file path returned) or failure, nor does it cover behavioral traits like idempotency or side effects. For a tool with no structured safety hints, this is inadequate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all parameters (songId, folder, filename) with clear descriptions. The description adds no additional meaning beyond what's in the schema, such as format details or examples. Baseline 3 is appropriate when the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Download') and resource ('a song MP3'), specifying the destination ('to a specified folder'). It distinguishes from siblings like 'suno_check_status' or 'suno_generate_song' by focusing on downloading rather than generating or monitoring. However, it doesn't explicitly differentiate from potential similar tools (e.g., if there were a 'suno_download_playlist'), so it's not a perfect 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention prerequisites (e.g., needing a generated song first), exclusions, or comparisons to siblings like 'suno_get_recent' for listing songs. Usage is implied from the action, but no explicit context is given.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate a song from a text description. Suno AI will write the lyrics.
| Name | Required | Description | Default |
|---|---|---|---|
| description | Yes | A description of the song you want, e.g. "An upbeat pop song about summer adventures" | |
| instrumental | No | Whether to generate an instrumental version | |
| model | No | Model to use | chirp-v4 |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It states the tool generates a song and Suno AI writes lyrics, but lacks critical details: it doesn't mention whether this is a synchronous or asynchronous operation (contrasting with 'suno_wait_for_songs'), what the output format is, potential rate limits, authentication requirements, or error conditions. For a generative AI tool with no annotation coverage, this is insufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise with just two sentences that directly state the tool's function. Every word earns its place, with no redundant or unnecessary information. It's front-loaded with the core purpose and efficiently adds one clarifying detail about lyrics.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of song generation (a creative AI task with potential async behavior) and the absence of both annotations and output schema, the description is incomplete. It doesn't explain what happens after generation, how to retrieve results, error handling, or relationship to sibling tools like 'suno_check_status' and 'suno_download_song'. For a tool in this context, more guidance is needed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all three parameters thoroughly. The description doesn't add any parameter-specific information beyond what's in the schema properties. The baseline score of 3 reflects adequate coverage via the schema alone, with no extra value from the description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Generate a song from a text description' specifies the verb (generate) and resource (song), and 'Suno AI will write the lyrics' adds detail about the AI's role. However, it doesn't explicitly distinguish this tool from sibling 'suno_generate_song', leaving some ambiguity about when to use each.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives like 'suno_generate_song' or other siblings. It mentions Suno AI writes lyrics, but doesn't specify prerequisites, exclusions, or contextual usage scenarios. The agent must infer usage from the name and parameters alone.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate a song with custom lyrics and style. Returns song IDs that can be used to check status.
| Name | Required | Description | Default |
|---|---|---|---|
| lyrics | Yes | The song lyrics. Can include structure tags like [Verse 1], [Chorus], [Bridge], etc. | |
| style | Yes | Musical style/genre tags, e.g. "indie folk, acoustic, male vocals, upbeat" | |
| title | No | Title for the song (optional) | |
| instrumental | No | Whether to generate an instrumental version (no vocals) | |
| model | No | Model to use: chirp-v4 (default), chirp-v3-5, chirp-crow (v5 beta) | chirp-v4 |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It mentions that the tool 'Returns song IDs that can be used to check status,' which hints at an asynchronous process, but doesn't disclose critical details like whether this is a long-running operation, rate limits, authentication requirements, error conditions, or what happens to generated songs over time.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately concise with two sentences that cover the core functionality and return value. It's front-loaded with the main purpose. However, the second sentence about checking status could be slightly more specific about the asynchronous nature of the operation.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a generative tool with 5 parameters, no annotations, and no output schema, the description is insufficient. It doesn't explain the asynchronous workflow (how status checking works), doesn't mention typical latency or limitations, and provides minimal guidance on parameter usage despite the schema having good coverage of technical details.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description mentions 'custom lyrics and style' which aligns with the two required parameters, but adds no additional semantic context beyond what's already in the schema (which has 100% coverage). It doesn't explain relationships between parameters, provide examples of effective style tags, or clarify how lyrics structure tags affect generation quality.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Generate a song with custom lyrics and style.' It specifies the verb 'generate' and resource 'song' with key inputs. However, it doesn't explicitly differentiate from sibling tools like 'suno_generate_from_description' which likely serves a similar purpose with different inputs.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It mentions that returns can be used to check status, but doesn't explain when to choose this over 'suno_generate_from_description' or how it relates to other siblings like 'suno_check_status' and 'suno_download_song' in a workflow.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get the current Suno account credits and subscription info
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It states the tool retrieves account information, implying a read-only operation, but doesn't disclose any behavioral traits such as authentication requirements, rate limits, error handling, or what the response format might look like. For a tool with zero annotation coverage, this is a significant gap in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, clear sentence: 'Get the current Suno account credits and subscription info.' It is front-loaded with the core purpose, has no unnecessary words, and efficiently conveys the essential information without waste, making it highly concise and well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (simple read operation) but lack of annotations and output schema, the description is incomplete. It doesn't explain what the return values include (e.g., credit balance, subscription details, format) or any behavioral aspects like error cases. For a tool with no structured data to supplement it, the description should provide more context to be fully helpful.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters, and the schema description coverage is 100%, meaning there are no parameters to document. The description doesn't need to add parameter semantics beyond what the schema provides. In such cases, the baseline score is 4, as there's no additional burden for parameter explanation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Get the current Suno account credits and subscription info.' It specifies the verb ('Get') and the resource ('Suno account credits and subscription info'), making it easy to understand what the tool does. However, it doesn't explicitly differentiate from sibling tools like 'suno_check_status' or 'suno_get_recent,' which might also retrieve account-related information, so it doesn't reach the highest score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention prerequisites, context for usage, or comparisons to sibling tools such as 'suno_check_status' or 'suno_get_recent.' This lack of explicit when-to-use or when-not-to-use information limits its helpfulness for an AI agent in selecting the correct tool.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get recently generated songs from your Suno account
| Name | Required | Description | Default |
|---|---|---|---|
| page | No | Page number (default: 1) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It states it retrieves songs but doesn't describe key traits like pagination behavior (implied by the 'page' parameter), rate limits, authentication needs, or what 'recently' means (e.g., time frame). This leaves significant gaps for a read operation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence that directly states the tool's purpose without unnecessary words. It is appropriately sized and front-loaded, making it easy to parse quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's low complexity (one optional parameter, no output schema), the description is minimally adequate. However, it lacks details on behavioral aspects like pagination or authentication, which are important for a read operation with no annotations, leaving room for improvement in completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, with the 'page' parameter documented as 'Page number (default: 1)'. The description adds no additional meaning beyond this, such as page size or ordering, so it meets the baseline of 3 where the schema handles parameter documentation adequately.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Get') and resource ('recently generated songs from your Suno account'), making the purpose understandable. However, it doesn't differentiate from sibling tools like 'suno_get_credits' or 'suno_check_status' which also retrieve information, so it lacks sibling distinction.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention scenarios like checking song status with 'suno_check_status' or retrieving credits with 'suno_get_credits', leaving the agent without context for selection among read operations.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Wait for songs to complete generation and return their URLs
| Name | Required | Description | Default |
|---|---|---|---|
| ids | Yes | Array of song IDs to wait for | |
| timeout | No | Maximum time to wait in seconds (default: 180) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions waiting for completion and returning URLs, but fails to describe key traits like whether it polls continuously, handles partial completions, or what happens on timeout. This leaves significant gaps in understanding the tool's behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence that directly states the tool's purpose without any wasted words. It is front-loaded and appropriately sized for its function, making it highly concise and well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of a waiting/polling tool with no annotations and no output schema, the description is insufficient. It doesn't explain the return format (e.g., structure of URLs, error handling), behavioral details like polling intervals, or how it integrates with sibling tools, leaving the agent with incomplete context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema description coverage is 100%, so the input schema already documents both parameters ('ids' and 'timeout') with descriptions and defaults. The description adds no additional meaning beyond what the schema provides, such as explaining the format of 'ids' or typical timeout values, resulting in a baseline score.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('wait for songs to complete generation') and the outcome ('return their URLs'), which is specific and actionable. However, it doesn't explicitly differentiate from sibling tools like 'suno_check_status' or 'suno_download_song', which might have overlapping functionality, so it misses the highest clarity mark.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives such as 'suno_check_status' for status checking or 'suno_download_song' for downloading. It lacks context on prerequisites (e.g., after song generation) or exclusions, leaving usage unclear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
suno_check_statussuno_download_songsuno_generate_from_descriptionsuno_generate_songsuno_get_creditssuno_get_recentsuno_wait_for_songsScored across 7 tools
Most tools have distinct purposes, but suno_generate_from_description and suno_generate_song could be confused as both generate songs, though one uses AI-written lyrics and the other custom lyrics. The other tools target clearly different operations like checking status, downloading, getting credits, and retrieving recent songs.
All tools follow a consistent snake_case pattern with the prefix 'suno_' and a clear verb_noun structure (e.g., check_status, download_song, generate_from_description). This predictability makes it easy for agents to understand and select tools.
With 7 tools, this server is well-scoped for managing Suno AI song generation, covering key operations like generation, status checking, downloading, and account management. Each tool earns its place without being overwhelming or insufficient.
The tool set covers the core lifecycle of song generation (create, check status, download, wait) and account management (credits, recent songs), but lacks operations for deleting or managing songs post-generation, which could be a minor gap for full lifecycle coverage.
Generate full songs with vocals, instrumentals and sound effects from text. One server for Suno, Google Lyria, Mureka, MiniMax, ACE-Step, YuE and ElevenLabs. Your agent can quote the credit cost before generating, start a song, wait for it to finish and return download links. Sign in with OAuth; optional daily credit caps; retries never charge twice.
Generate Suno AI music (v5.5) from any MCP client. Async; billed only on success.
MCP server for Suno AI music generation, lyrics, and covers
Make songs, change one part's lyrics in the same voice, and reimagine songs in new styles.
