Skip to main content
Glama
xaxeane

suno-v6-kie-mcp

by xaxeane
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Suno V6 Kie.ai MCP

MCP server ringan untuk membuat lagu atau musik menggunakan Suno V6, V6_MINI, dan V6_WILD melalui API resmi Kie.ai. Server mendukung stdio untuk MCP client lokal dan Streamable HTTP untuk Manus Custom MCP Connector.

Fitur

  • suno_generate_music untuk membuat lagu vokal atau instrumental.

  • suno_get_task_status untuk mengambil status dan URL audio hasil generate.

  • Dukungan V6, V6_MINI, dan V6_WILD.

  • Dukungan duration 10–360 detik pada custom mode.

  • Tidak menyimpan API key atau hasil audio ke database.

  • HTTP endpoint /mcp dan health check /health.

Related MCP server: Suno Autopilot MCP

Persiapan

Gunakan Node.js 20 atau lebih baru dan API key dari Kie.ai API Key.

git clone https://github.com/xaxeane/suno-v6-kie-mcp.git
cd suno-v6-kie-mcp
npm install
npm run build

Menjalankan sebagai MCP lokal (stdio)

KIE_AI_API_KEY="YOUR_KIE_API_KEY" npm start

Contoh konfigurasi MCP client yang mendukung command lokal:

{
  "mcpServers": {
    "suno-v6-kie": {
      "command": "node",
      "args": ["/absolute/path/to/suno-v6-kie-mcp/dist/index.js"],
      "env": {
        "KIE_AI_API_KEY": "YOUR_KIE_API_KEY"
      }
    }
  }
}

Menjalankan sebagai remote MCP untuk Manus

Remote deployment diperlukan karena Manus Custom MCP Connector membutuhkan endpoint HTTPS yang dapat diakses Manus.

KIE_AI_API_KEY="YOUR_KIE_API_KEY" \
KIE_MCP_HTTP_TOKEN="GENERATED_MCP_TOKEN" \
MCP_TRANSPORT=http \
MCP_HTTP_HOST=0.0.0.0 \
MCP_HTTP_PORT=3000 \
MCP_ALLOWED_HOSTS="mcp.example.com" \
npm start

Gunakan reverse proxy/TLS sehingga endpoint publik menjadi:

https://mcp.example.com/mcp

Set header autentikasi pada connector Manus:

Authorization: Bearer GENERATED_MCP_TOKEN

KIE_AI_API_KEY hanya disimpan di environment server dan tidak dikirim ke Manus.

Docker

docker build -t suno-v6-kie-mcp .
docker run -d --name suno-v6-kie-mcp \
  -p 3000:3000 \
  -e KIE_AI_API_KEY="YOUR_KIE_API_KEY" \
  -e KIE_MCP_HTTP_TOKEN="GENERATED_MCP_TOKEN" \
  -e MCP_TRANSPORT=http \
  -e MCP_HTTP_HOST=0.0.0.0 \
  -e MCP_ALLOWED_HOSTS="mcp.example.com" \
  suno-v6-kie-mcp

Setelah service aktif, buka Manus → Connectors → Add connector → Custom MCP, masukkan URL /mcp, tambahkan header Bearer token, lalu aktifkan connector.

Contoh penggunaan

{
  "prompt": "Cinematic Indonesian pop song about chasing dreams under the city lights",
  "customMode": false,
  "instrumental": false,
  "model": "V6"
}

Untuk custom mode:

{
  "prompt": "Langit malam menjadi saksi langkah kita\nTakkan menyerah mengejar cahaya",
  "customMode": true,
  "instrumental": false,
  "model": "V6_WILD",
  "style": "Indonesian cinematic pop, emotional female vocal, orchestral chorus",
  "title": "Mengejar Cahaya",
  "duration": 180
}

Setelah menerima task_id, panggil suno_get_task_status dengan:

{ "taskId": "TASK_ID_DARI_RESPONSE" }

Status yang umum: waiting, queuing, generating, success, atau fail. URL audio tersedia pada result ketika task berhasil.

Environment variables

Variable

Wajib

Keterangan

KIE_AI_API_KEY

Ya

Bearer API key Kie.ai

KIE_AI_BASE_URL

Tidak

Default https://api.kie.ai

MCP_TRANSPORT

Tidak

http untuk remote MCP; default stdio

MCP_HTTP_HOST

Tidak

Default 127.0.0.1

MCP_HTTP_PORT

Tidak

Default 3000

KIE_MCP_HTTP_TOKEN

Remote

Bearer token untuk client MCP

MCP_ALLOWED_HOSTS

Remote

Host/domain publik yang diizinkan

Keamanan

Gunakan HTTPS pada deployment publik, token MCP acak, dan jangan commit file .env atau API key. Kie.ai mengenakan biaya/kredit sesuai akun dan model yang digunakan; server ini tidak melakukan approval billing di luar alur MCP client.

Lisensi

MIT

Available Tools

2 tools
suno_generate_musicB

Generate a song or instrumental track using Kie.ai Suno V6, V6_MINI, or V6_WILD. The returned task_id can be checked with suno_get_task_status.

ParametersJSON Schema
NameRequiredDescriptionDefault
modelNoV6
styleNoGenre, mood, or production style. Required in custom mode.
titleNoTrack title. Required in custom mode.
promptYesLyrics in custom mode, or a music idea in non-custom mode.
durationNoSeconds; custom mode only for V6 family.
personaIdNo
customModeYesUse advanced controls. Style and title are required when true.
audioWeightNo
callBackUrlNo
styleWeightNo
vocalGenderNo
instrumentalYesGenerate without vocals.
negativeTagsNo
personaModelNo
weirdnessConstraintNo

TDQS

B3.1/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It discloses that the tool returns a task_id for async checking, which is useful, but it does not mention that generation is asynchronous, that it may take time, that it costs credits, or that it creates a remote resource. The description is thin for a generation tool with 15 parameters.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the core action and model family, and the task_id handoff is a useful addition. No wasted words, though it could have used the space to explain async behavior or key parameter relationships.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex generation tool with 15 parameters, no annotations, and no output schema, the description is incomplete. It does not explain the async nature, the customMode vs non-customMode distinction, required fields in custom mode, or what the response contains beyond task_id. The sibling handoff is helpful but not sufficient.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is only 40%, so the description should compensate. It does not explain any parameters beyond the model family and the task_id flow. The schema covers some fields (prompt, customMode, instrumental, style, title, duration), but many parameters (audioWeight, styleWeight, vocalGender, negativeTags, personaId, personaModel, weirdnessConstraint, callBackUrl) are undocumented in both schema and description. The description adds minimal value over the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb ('Generate') and resource ('song or instrumental track using Kie.ai Suno V6, V6_MINI, or V6_WILD'), and it names the sibling tool for checking the returned task_id. It is clear what the tool does, though it doesn't explicitly contrast with the sibling beyond the task_id handoff.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage by saying the returned task_id can be checked with suno_get_task_status, which gives a follow-up workflow. However, it does not state when to use this tool vs alternatives, nor does it mention that customMode requires style and title (that's only in the schema).

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

suno_get_task_statusA

Get the current status and generated audio URLs for a Kie.ai Suno task.

ParametersJSON Schema
NameRequiredDescriptionDefault
taskIdYes

TDQS

A3.8/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the behavioral burden. It does disclose that the call returns status and audio URLs, which implies a read-only polling operation, but it does not cover what happens for pending tasks, invalid task IDs, or whether repeat calls are safe.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single, front-loaded sentence with no filler. Every word contributes to identifying the operation, the resource, and the expected return payload.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple one-parameter status retrieval tool, the description names the main return data and is largely sufficient for correct invocation. It would be more complete with explicit guidance about post-generation usage or behavior when the task is still processing, but those are not essential for a basic call.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema has no property descriptions and schema description coverage is 0%, but the single parameter taskId is self-explanatory given the tool name and description. The description does not add meaning beyond the parameter name, but the gap is minor because there is only one required parameter.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb ('Get'), the resource ('current status'), and the key output ('generated audio URLs') for a Kie.ai Suno task. This clearly distinguishes the tool from the sibling suno_generate_music, whose purpose is creation rather than retrieval.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies the tool should be used after a task exists, since it fetches status and results for an already-created task. However, it does not explicitly say 'use after suno_generate_music' or explain when polling is appropriate, leaving the usage context to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 2 tool updatesv1.0.0
    • First observedsuno_generate_music
    • First observedsuno_get_task_status
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.7/5.0

Scored across 2 tools

Disambiguation5/5

The two tools have clearly distinct purposes: one initiates music generation and returns a task ID, while the other retrieves status and generated audio URLs for that task. There is no overlap or ambiguity between them.

Naming Consistency5/5

Both tool names follow the same suno_ prefix with a clear verb_noun pattern: generate_music and get_task_status. The naming is consistent, descriptive, and predictable.

Tool Count4/5

With only two tools, the server is minimal but appropriately scoped for an asynchronous generation workflow: submit a task and check its result. It is slightly thin, but each tool serves an essential role.

Completeness4/5

The core lifecycle of submitting a generation task and polling for results is covered. Minor gaps exist, such as no cancel or list operations, but for a focused Suno generation server these are not critical.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers