producer-music — Claude Code skill: music and install
producer-music creates AI music through the AceDataCloud Producer API, including songs, lyrics, covers, and instrumentals. It also edits existing tracks and separates stems.
Apache-2.0 236 installs ★ 17 updated 2026-11-07
npx skills add AceDataCloud/Skills --skill producer-music -a claude-codeWhat producer-music does
This skill connects Claude Code to AceDataCloud’s Producer API for music generation and editing. It can start a song from a prompt or custom lyrics, write lyrics, create a cover or variation, continue a track from a timestamp, replace a section, swap vocals or instrumentals, and split audio into stems.
Generation requests are asynchronous. A returned task ID must be checked through POST /producer/tasks until the task completes, rather than sending another generation request. The skill also exposes workflows for WAV and video output plus reference-audio upload.
Set ACEDATACLOUD_API_TOKEN in ~/.openclaw/.env or in the shell environment. Custom generation can use lyric, instrumental, a model such as FUZZ-2.0 Pro, and controls for lyrics strength, sound strength, randomness, or reproducibility. Without the API token, AceDataCloud requests cannot run.
What it helps with
- create a song from a written prompt
- draft lyrics for a new track
- continue audio from a chosen timestamp
- replace a section of an existing song
- split a track into separate stems
Example prompts for the agent
- Create an upbeat electronic track with synth leads.
- Write lyrics about stargazing on a summer night, then generate a song from them.
- Continue
existing-audio-idfrom30.792and add a new verse.
How to install producer-music
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill producer-music -a claude-code2Manually: put the files in place
~/.claude/skills/producer-music/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/producer-music/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill producer-music -a codex2Manually: put the files in place
.agents/skills/producer-music/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill producer-music -a cursor2Manually: put the files in place
.cursor/skills/producer-music/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install AceDataCloud/Skills skills/producer-music2Manually: put the files in place
.github/skills/producer-music/SKILL.md1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill producer-music -a gemini-cli2Platform CLI
gemini skills install https://github.com/AceDataCloud/Skills --path skills/producer-musicManually: put the files in place
.devin/skills/producer-music/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill producer-music -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about producer-music
How do I install producer-music in Claude Code?
Run npx skills add AceDataCloud/Skills --skill producer-music -a claude-code, then check ~/.claude/skills/producer-music/SKILL.md.
What does producer-music do?
It uses the Producer API to create songs and lyrics, make covers, extend tracks, replace sections, swap vocal or instrumental parts, and separate stems.
What does producer-music need to run?
Set ACEDATACLOUD_API_TOKEN in ~/.openclaw/.env or the shell environment. Generation is asynchronous, so poll POST /producer/tasks with the returned task ID.
Source text
SKILL.md · Apache-2.0
---
name: producer-music
description: Generate AI music with Producer via AceDataCloud API. Use when creating songs, generating lyrics, extending tracks, creating covers, swapping vocals/instrumentals, replacing song sections, or uploading reference audio. Supports custom lyrics, instrumental-only mode, and multiple creative actions.
license: Apache-2.0
metadata:
author: acedatacloud
version: "1.0"
compatibility: Requires ACEDATACLOUD_API_TOKEN in .env file (get an API key at https://platform.acedata.cloud/console/applications).
---
# Producer Music Generation
Generate AI music through AceDataCloud's Producer API.
## First-time setup
1. Install the official Skill with `clawhub install @acedatacloud/producer`. Check that the ClawHub publisher is **@acedatacloud**.
2. Sign in at [AceDataCloud → Applications](https://platform.acedata.cloud/console/applications), open **General application**, and check Producer access, balance, and current pricing. Copy its API key, or select **Manage Keys → Create** for a separate key. Copy the key value without `Bearer `; a `platform-...` management token is not a service API key.
3. If you enable **Allowed APIs** restrictions, include `/producer/audios` and `/producer/tasks`. Advanced workflows below may call additional routes.
4. On the OpenClaw Gateway host, put `ACEDATACLOUD_API_TOKEN=your_api_key_here` in `~/.openclaw/.env`, then restart the Gateway and start a new session. For a direct terminal test, run `export ACEDATACLOUD_API_TOKEN="your_api_key_here"` in that shell. Never put the key in a prompt or commit an `.env` file.
5. Run the **Quick Start** below. If it returns a task ID, query that same task until it finishes; do not submit a second billable generation just to check progress.
[Illustrated key setup, first run, and troubleshooting](https://github.com/AceDataCloud/Skills#first-run-with-clawhub-and-openclaw) · [Token details](https://github.com/AceDataCloud/Skills/blob/main/skills/_shared/authentication.md)
## Quick Start
```bash
curl -X POST https://api.acedata.cloud/producer/audios \
-H "Authorization: Bearer $ACEDATACLOUD_API_TOKEN" \
-H "Content-Type: application/json" \
-d '{"action": "generate", "prompt": "upbeat electronic dance track with synth leads"}'
```
> **Async:** All generation is async. See [async task polling](https://github.com/AceDataCloud/Skills/blob/main/skills/_shared/async-tasks.md). Poll via `POST /producer/tasks` with `{"id": "..."}` every 3-5 seconds.
## Models
| Model | Notes |
|-------|-------|
| `FUZZ-2.0 Pro` | Default, highest quality |
| `FUZZ-2.0` | Standard quality |
| `FUZZ-2.0 Raw` | Raw output variant |
| `FUZZ-1.1 Pro` | Pro v1.1 |
| `FUZZ-1.0 Pro` | Pro v1.0 |
| `FUZZ-1.0` | v1.0 |
| `FUZZ-1.1` | v1.1 |
| `FUZZ-0.8` | Legacy |
## Actions
| Action | Description |
|--------|-------------|
| `generate` | Create a new song from prompt or custom lyrics |
| `cover` | Create a cover version of an existing song |
| `extend` | Continue a song from a specific timestamp |
| `replace_section` | Replace a time range in an existing song |
| `swap_vocals` | Extract and swap vocal tracks |
| `swap_instrumentals` | Extract and swap instrumental tracks |
| `variation` | Generate a variation of an existing song |
| `stems` | Separate a song into stems |
## Choose inputs by action
Always send `action`, then supply only the inputs needed for that operation:
| Operation | Required inputs |
|---|---|
| `generate`, `custom=false` (default) | Nonempty `prompt` |
| `generate`, `custom=true`, `instrumental=false` | Nonempty `lyric` |
| `generate`, `custom=true`, `instrumental=true` | Neither lyrics nor an extra prompt is required |
| `extend` | Source `audio_id`, `continue_at > 0`, and nonempty `lyric` unless `instrumental=true` |
| `cover`, `variation`, `replace_section`, `swap_vocals`, `swap_instrumentals`, `stems` | Source `audio_id`; add the selected operation's editing controls |
Do not insert empty lyrics or prompts just to satisfy a universal required-field rule. In MCP, supplying `lyric` to `producer_extend_music` or `producer_replace_section` selects custom lyrics mode; instrumental extension can omit lyrics with `instrumental=true`.
## Workflows
### 1. Generate from Prompt
```json
POST /producer/audios
{
"action": "generate",
"prompt": "chill lo-fi hip hop with rain sounds and soft piano"
}
```
### 2. Custom Lyrics Mode
```json
POST /producer/audios
{
"action": "generate",
"custom": true,
"title": "Midnight City",
"lyric": "[Verse]\nNeon lights reflect on wet streets\n[Chorus]\nMidnight city never sleeps",
"prompt": "nocturnal electronic pop",
"instrumental": false
}
```
### 3. Instrumental Only
```json
POST /producer/audios
{
"action": "generate",
"custom": true,
"instrumental": true
}
```
### 4. Extend Song
```json
POST /producer/audios
{
"action": "extend",
"audio_id": "existing-audio-id",
"custom": true,
"lyric": "[Verse]\nA new chapter starts tonight",
"continue_at": 30.792
}
```
### 5. Replace Section
```json
POST /producer/audios
{
"action": "replace_section",
"audio_id": "existing-audio-id",
"custom": true,
"lyric": "[Chorus]\nNew words for this section",
"replace_section_start": 15,
"replace_section_end": 30
}
```
### 6. Separate into Stems
```json
POST /producer/audios
{
"action": "stems",
"audio_id": "existing-audio-id"
}
```
### 7. Generate Lyrics
```json
POST /producer/lyrics
{
"prompt": "a love song about stargazing on a summer night"
}
```
### 8. Get WAV / Video
```json
POST /producer/wav
{"audio_id": "your-audio-id"}
POST /producer/videos
{"audio_id": "your-audio-id"}
```
### 9. Upload Reference Audio
```json
POST /producer/upload
{
"audio_url": "https://example.com/reference.mp3"
}
```
## Parameters
| Parameter | Type | Description |
|-----------|------|-------------|
| `action` | string | Required for every `/producer/audios` request; see actions table |
| `prompt` | string | Required for non-custom `generate`; not unconditionally required for editing or custom instrumental generation |
| `model` | string | Model (e.g., `"FUZZ-2.0 Pro"`) |
| `custom` | boolean | Enable custom lyrics mode |
| `instrumental` | boolean | Pure instrumental (no vocals) |
| `title` | string | Song title |
| `lyric` | string | Required for custom non-instrumental `generate` and non-instrumental `extend`; use `[Verse]`, `[Chorus]` tags |
| `audio_id` | string | Existing audio ID (for edit actions) |
| `continue_at` | number | Required for `extend`; seconds greater than 0, including fractional seconds |
| `replace_section_start` | number | Start time of section to replace |
| `replace_section_end` | number | End time of section to replace |
| `lyrics_strength` | 0-1 | Lyrics adherence (default: 0.5) |
| `sound_strength` | 0.2-1 | Sound quality weight (default: 0.5) |
| `weirdness` | 0-1 | Creative randomness (default: 0.5) |
| `seed` | string | Seed for reproducibility |
## Response Structure
```json
{
"data": [
{
"id": "audio-id",
"audio_url": "https://cdn.example.com/song.mp3",
"video_url": "https://cdn.example.com/video.mp4",
"image_url": "https://cdn.example.com/cover.jpg",
"title": "Song Title",
"lyric": "full lyrics...",
"style": "electronic, dance",
"model": "FUZZ-2.0 Pro"
}
]
}
```
## Gotchas
- Use `[Verse]`, `[Chorus]`, `[Bridge]`, `[Outro]` tags in custom lyrics
- `continue_at` is in **seconds** — the song extends from that point
- `replace_section_start` / `replace_section_end` define the time range to regenerate
- `weirdness` at 0 = predictable, at 1 = highly experimental
- Upload a reference audio via `/producer/upload` to get an audio ID for use with `cover` or `extend`
- WAV and video downloads are separate endpoints — call them after the main generation completes
- **CRITICAL:** Check the `state` field in task responses — only `state: "complete"` with `success: true` means done. During `pending`, the API may return intermediate `audio_url` values (streaming previews). Do NOT stop polling just because `audio_url` is non-empty
© acedatacloud · Apache-2.0 · AceDataCloud/Skills