Skip to content

106 social media skills for AI agents - strategy, writing, video, design, platform growth, publishing, and analytics. Works with Claude, Cursor, OpenClaw, Hermes & 40+ agents.

4.2 KB
Raw Blame

Suno — integration & connection guide

What it is: the AI music generator (v5.5 + Suno Studio) this stack uses for owned brand sound — themes, stings, scores, jingles. Suno generates; the human's ears judge; the track scores the edit; WoopSocial publishes the content. Skill: skills/suno/; the router is skills/ai-music-and-sound/. The legal terms here changed materially in late 2025 and keep moving — verify-quarterly against suno.com's live Terms/Help Center; treat pre-Warner-deal reviews as stale on rights.

Connection layers

  1. API (2025 release; Python/Node libraries; Zapier/Make integrations): the agent can drive generation where a key/connection exists — tier gating for API access conflicts across sources (Enterprise-only vs paid-tier reports): verify before promising automation. No auto-publishing of unheard audio — the human's ears approve everything.
  2. In-app (human executes): the agent supplies the brief (genre + mood anchor + instruments + vocal spec), meta-tag structure, exclude-styles, lyrics, the consistency plan, and the archive checklist; the human generates, curates, and exports.
  3. Handoff: WAV/MP3 exports → capcut (scoring the edit — replacing its license-trap library music) or podcast-and-audiograms (stings) → the finished content → scheduling-and-queue → WoopSocial: upload media → attach → validate → create post (see tools/integrations/woopsocial.md). WoopSocial does not generate audio or mix it into video — score the edit before upload; trending audio is native-only.

Tiers (≈; verify in-app)

Free: 50 credits/day (10 songs) — strictly non-commercial, forever (no retroactive rights after upgrading; regenerate keepers on a paid plan). **Pro ≈ $8/mo annual ($10 monthly):** 2,500 credits (~500 songs), commercial rights, Personas, Voices, stems, uploads. Premier ≈ $24/mo annual ($30): 10,000 credits

  • Studio (browser DAW: multitrack, 12 zero-bleed stems, MIDI export) — same legal rights as Pro; the extra buys volume + tooling. Credits don't roll over; top-ups persist but need an active sub.

The rights spine (the reason this file exists — attribute; verify-quarterly)

  • Litigation live: Warner settled + partnered (Nov 2025); UMG + Sony still suing (fair-use summary-judgment hearing July 2026; GEMA verdict in Germany mid-2026). No indemnification — commercial rights don't include legal defense; favor generic style language over artist references.
  • Post-Warner terms: "you own it" → granted commercial rights (Suno remains functionally the author); free commercial downloads removed; paid download caps described as forthcoming; pre-deal models scheduled for deprecation when licensed models ship — archive WAVs/stems/MIDI now; artist opt-in (compensated, artist-controlled) is the consent path for artist voices — never prompting or cloning.
  • Protectability: purely-AI music isn't copyrightable (prompting ≠ authorship); human elements (original lyrics, re-recorded parts over stems, DAW arrangement) strengthen claims and reclassify as AI-assisted for distributors.
  • Disclosure: DDEX AI flag required at distribution (Spotify/Apple enforcement since late 2025); no PRO registration (ASCAP/BMI) for pure-AI works; AI-disclosure on content where platform/region requires (EU AI Act).

Known frictions (community-reported; attribute)

Post-May-2026 Persona/Voice regressions (voices reverting to genre defaults — test before client promises) · Studio's failed stem regenerations still deducting credits (batch deliberately; export to a DAW) · ISRC self-upload blocking on users' own distributed tracks · performance honesty: one firm's data shows fully-AI tracks underperform human releases on saves ~25–40%.

Hard lines

Human ears judge every track · paid tier before anything commercial · never clone a real artist's voice or fake a collab · litigation-aware prompting (generic styles) · DDEX AI flag at distribution; no PRO registration for pure-AI · archive keepers ahead of deprecation · never state a tier/legal status as immutable fact · WoopSocial does not generate, mix, or license audio; trending audio is native-only.