calesthio/generative-media-skills/suno-music

Suno music generation and rights advisory

Suno is a text-to-music service that generates full songs (vocals + instrumental) from a style description and optional lyrics, plus editing tools (extend, cover, remaster, stem separation, a DAW-style Studio). This skill covers two things a production agent must get right at the same time: making good music with the current models and advising accurately on the rights, licensing, and legal reality of using that music commercially.

The rights half is not an appendix. As of mid-2026 the single most common way to give a production team bad advice is to treat Suno output as freely ownable, copyrightable, and API-accessible. Read the "Rights, licensing, and legal reality" section before advising on any commercial or distribution use.

Evidence labels used below

• [Documented] — stated in Suno's own docs, terms, help center, or blog, or in a

primary legal/press source, with a URL in Sources.

• [First-party claim] — Suno's own marketing/announcement language, reported as a

claim rather than an independently verified fact.

• [Heuristic] — production practice inferred from how the tools behave; useful defaults,

not guarantees.

Volatility warning. Models, tiers, prices, terms, litigation status, and API access change fast. Every volatile fact below carries a verification date of 2026-07-10 unless noted. Re-verify anything legal or version-specific against suno.com before relying on it.

1. Current models and what they do

[Documented] Model timeline (Suno help center "Model Timeline", verified 2026-07-10):

• [Documented] V5 launched ~September 23–25, 2025 and is available only to paid (Pro/Premier)

subscribers; free-tier users generate on older models. Suno also shipped Suno Studio (a generative audio workstation with a multitrack editor and MIDI/stem export) on the Premier plan around Sep 25, 2025.

• [Documented] V5.5 (Mar 26 2026) added three features: Voices (capture or upload your

own voice, sing on your creations; includes voice verification, recordings kept private), Custom Models (Pro/Premier can train up to three personal model versions on their own catalog), and My Taste (preference personalization, available to all tiers).

• [First-party claim] Suno marketed V5 as its "most powerful" / "world's best music model"

and V5.5 as "more expressive." Treat superlatives as marketing, not benchmarks — there is no neutral, published head-to-head that establishes Suno as a universal "best." When a user asks "is Suno the best," answer on fitness for their use, not on the marketing claim.

• [Documented — anticipated] Under the Warner deal (see §5), Suno has said higher-quality

models trained on licensed WMG recordings are forthcoming. As of 2026-07-10 no such licensed model is confirmed shipped under a public name; treat "WMG-licensed model" as announced-but-unshipped and re-verify.

Version selection heuristic. For any track intended for release, generate on V5/V5.5 (paid). Free-tier output is both lower quality and non-commercial (§4), so it is only appropriate for scratch ideas, learning, and personal listening.

2. The generation surface: fields and controls

Suno's create screen (Custom mode) has two text inputs that do different jobs — mixing them up is the most common cause of bad output.

• Style / description field — genre, subgenre, mood, tempo feel, instrumentation, production

character, vocal type, era, reference adjectives. This is where the sonic identity lives.

• Lyrics field — the actual words plus structural metatags in square brackets. If you put

genre words here they may be sung. If you leave it empty (or use "Instrumental"), you get an instrumental.

• [Documented] Exclude Styles (negative prompt) — in Custom mode under Advanced Options; list

what you do not want (e.g. a genre, an instrument, "male vocals") and Suno deprioritizes it.

• Simple vs Custom mode — Simple mode takes one free-text prompt and writes lyrics/structure

for you; Custom mode exposes the style field, lyrics field, exclude-styles, model picker, and weirdness/style-influence sliders. Use Custom for anything you want control over.

[Heuristic] Style field craft

• Lead with the anchor: one clear genre + one clear mood, then layer detail. `melodic

drum-and-bass, euphoric, driving` beats a 40-word soup of adjectives.

• Name instruments and vocal type explicitly when they matter (female alto, breathy,

analog synth bass, brushed jazz drums).

• Tempo: you can state BPM in the style field, or use a [bpm-128] metatag in lyrics; giving a

numeric BPM is more reliable than "fast."

• Don't over-stuff. Beyond ~6–10 strong descriptors, added words dilute rather than sharpen.
• Contradictions (e.g. "minimal, wall-of-sound maximalist") produce mush; pick a lane.

3. Lyrics, structure, and metatags

[Documented] Structural metatags go in the lyrics field inside square brackets and steer arrangement, dynamics, and vocal behavior. Without them Suno infers structure from line breaks; with them you remove the guesswork.

Common, reliably-understood tags ([Documented] for the core structure tags; [Heuristic] for exact behavior, which varies by model):

[Intro]
[Verse]        [Verse 1] [Verse 2]
[Pre-Chorus]
[Chorus]
[Post-Chorus]
[Bridge]
[Instrumental]  [Guitar Solo]  [Drop]
[Outro]         [Fade Out]
[no-vocals]     [bpm-128]  [key-C-minor]  [time-signature-3/4]

[Heuristic] Lyric and structure practice

• [Chorus] triggers chorus behavior — bigger arrangement, hook, repetition on later

occurrences. Reusing the exact same chorus text each time yields a consistent, memorable hook; varying it slightly yields evolution.

• Keep tags short (1–3 words). Long descriptive tags are less reliable than a short structural

tag plus a style-field descriptor.

• To force an instrumental, put [Instrumental] / [no-vocals] and leave lyrics empty, or select

the Instrumental toggle.

• Inline performance cues (whispered, ad-lib, spoken) work inconsistently across models — test,

don't assume.

• Suno can auto-write lyrics; for a controlled brief, write your own lyrics so you also own that

copyright layer (§4).

4. Editing, iteration, and stems

• Extend — continue a track past its current end; useful for building full arrangements from a

strong 1–2 min section.

• Cover — re-perform existing lyrics/melody in a new style.
• Remaster — [Documented] re-render an older track on a newer model (e.g. lift a V4 track to

V5 quality), with control over how much it varies from the original.

• Add Vocals / Add Instrumental — layer onto an existing track (V4.5+ onward).
• Personas / Voices — [Documented] Persona captures the vocal/stylistic identity of a prior

generation to carry it across new songs; V5.5 Voices lets you record or upload your own voice to sing on creations (with verification; recordings private by default). Do not use a Voice/Persona to imitate a real artist you have no rights to (§5, §6).

• Stems — [Documented / First-party claim] Suno can split a generation into individual

stems (vocals, drums, bass, and more — marketed as up to ~12 stems) and Suno Studio (Premier) exports multitrack stems and MIDI for finishing in a DAW (Ableton, Logic, FL, etc.).

[Heuristic] Iteration workflow. Generate several candidates on a tight style prompt → pick the best structural take → Extend to full length → Remaster/clean if needed → export stems for mix and master in a DAW. Regenerating from a slightly varied prompt is usually faster than fighting one bad take with edits.

5. Rights, licensing, and legal reality (read before advising on commercial use)

This section is the core of the skill. Get the tier logic and the copyright caveat right every time.

5.1 Ownership and commercial use by tier

[Documented] (Suno Terms of Service, last revised 2026-03-26; Suno help center "Rights & Ownership"; verified 2026-07-10):

• Free / Basic tier: Suno retains ownership. You get a limited license to use output for

lawful, internal, personal, non-commercial purposes only, and you must give attribution to Suno. No distribution, no monetization, no commercial use.

• Pro / Premier (paid) tiers: Suno assigns to you all of its right, title, and interest in

the output generated from your submissions while you were subscribed. You may distribute, monetize, and use it commercially. Premier gets the same commercial rights as Pro — the extra cost buys volume (credits) and tooling (Studio), not broader legal rights.

• Timing is what matters, not current status: ownership/commercial rights attach based on your

subscription at the moment the track was generated. [Documented] Songs made while paid stay commercial-usable even after you cancel. Conversely, upgrading later does not retroactively grant rights to a track you made on the free tier — you must regenerate it while paid.

Approximate 2026 pricing (verify — volatile): Pro ≈ $10/mo (~$8 annual), ~2,500 credits; Premier ≈ $30/mo (~$24 annual), ~10,000 credits + Suno Studio. Prices/credits change; confirm on suno.com before quoting.

5.2 The copyright caveat that ownership does not cure

[Documented] Two separate things, and agents routinely conflate them:

1. Contractual ownership from Suno — paid tiers get this (Suno assigns its interest to you).
2. Copyright protection under law — a different question. Suno's terms state it makes **no

representation or warranty that any copyright will vest in any Output**. Suno's own help center warns output "may not be eligible for copyright protection."

Why: U.S. copyright requires human authorship; the U.S. Copyright Office has repeatedly held that material generated by AI from a prompt is not protectable, because writing a prompt is not authorship of the resulting expression. [Documented] Practical consequence: a fully Suno-generated track can be owned (contractually, on a paid tier) and legally used commercially, yet not copyrightable, meaning you may have limited ability to stop others from copying it. [Heuristic] The strongest human-authorship hook is writing your own lyrics (you own that literary copyright independently) and doing substantial human arrangement/editing; this improves — but does not guarantee — registrability. This varies by country; direct consequential decisions to a qualified attorney or the relevant copyright office.

5.3 Distribution, monetization, and disclosure