Generate AI music with Suno via API or browser, with prompt engineering and song extensions.
On first use, read setup.md for integration guidelines.
User wants to generate music with Suno. Agent can use hosted APIs for programmatic generation, browser automation for direct platform interaction, or guide prompt engineering for manual use.
Memory at ~/suno/. See memory-template.md for structure.
~/suno/
├── [memory.md] # Created on first use: preferences, successful prompts
├── [projects/] # Per-project song tracking
└── [songs/] # Downloaded audio files
| Topic | File |
|---|---|
| Setup | setup.md |
| Memory | memory-template.md |
| API usage | api.md |
| Browser automation | browser.md |
| Prompt crafting | prompts.md |
| Style tags | styles.md |
| Lyrics guide | lyrics.md |
| Situation | Method |
|---|---|
| Programmatic generation | Hosted API (aimusicapi.ai, EvoLink) |
| Visual interaction | Browser at suno.com |
| Just need prompts | Prompt engineering only |
[genre] [subgenre] [mood] [instruments] [voice] [era/influence]
Example: "indie folk melancholic acoustic guitar soft female vocals 90s"
[Verse]
Your lyrics here
[Chorus]
Hook section
[Bridge]
Contrast
[Outro]
Ending
Suno generates clips. Build full songs:
All APIs follow: generate → poll for completion → retrieve audio URL.
Generation takes 30-90 seconds. See api.md for code examples.
Two main options for programmatic generation:
aimusicapi.ai — Get API key at aimusicapi.ai EvoLink — Get API key at evolink.ai
Both provide REST APIs for generation, custom lyrics, and extensions.
See api.md for detailed code examples and endpoint documentation.
# Conceptual flow (see api.md for real code)
1. POST /generate with prompt
2. Receive task_id
3. Poll /task/{id} every 5 seconds
4. Get audio_url when status="completed"
When API isn't available or user prefers visual interaction:
See browser.md for detailed automation steps.
| Genre | Pattern |
|---|---|
| Electronic | electronic [subgenre] [mood] synth [texture] |
| Rock | [sub]rock [energy] [guitars] [vocals] [decade] |
| Pop | pop [mood] [tempo] [vocals] [production] |
| Hip Hop | hip hop [subgenre] [beat] [flow] [era] |
soft female vocals, ethereal, breathy
deep male vocals, baritone, raspy
instrumental, no vocals
See prompts.md and styles.md for comprehensive guides.
| Trap | Problem | Solution |
|---|---|---|
| Vague prompts | Random output | Be specific with genre, mood |
| Contradictions | Confuses model | Consistent descriptors |
| Too many keywords | Dilutes focus | 8-12 key terms max |
| No structure tags | Awkward lyrics | Use [Verse], [Chorus] |
This skill creates ~/suno/ on first use:
All data stays local. API keys should be stored as environment variables.
This skill does:
This skill does NOT:
~/suno/When using hosted APIs, requests go to:
| Endpoint | Data Sent | Purpose |
|---|---|---|
| api.aimusicapi.ai | Prompts, lyrics | Music generation |
| api.evolink.ai | Prompts, lyrics | Music generation |
| suno.com | Browser session | Direct platform access |
API keys authenticate requests. Prompts and lyrics are sent for processing.
By using this skill with APIs, prompts and lyrics are sent to third-party services for music generation. Only use services you trust with your creative content.
Install with clawhub install <slug> if user confirms:
audio — Audio processing and editingvideo — Combine music with video contentffmpeg — Audio format conversionclawhub star sunoclawhub sync