SuperAPI AI Directory

guided-meditation-composer

Design & creation

View repository

Updated 10-15 21:46

About this skill

Guided Meditation Composer

Choose the production lane first

Preferred professional lane: separate narration + instrumental bed.

  • Write the spoken script for a human or TTS voice.
  • Generate an instrumental Suno bed with no vocals.
  • Mix later with enough spectral and dynamic space for speech.

Experimental lane: Suno spoken word.

  • Put the script in Custom Mode with [Spoken Word], sparse ambient direction, and short phrases.
  • Expect less reliable pacing and diction. Label the result experimental.

Do not turn a meditation into a sung song unless the user asks for one.

Safety boundaries

This is creative wellness content, not diagnosis or treatment.

  • Never promise healing, nervous-system regulation, trauma release, manifestation certainty, or clinical outcomes.
  • Do not use forceful breath holds, hyperventilation, deprivation, pain endurance, or "push through" language.
  • Prefer normal breathing. If changing breath, make it gentle and optional.
  • Use permissive phrasing: "if it feels comfortable", "you might notice", "let your eyes close or soften".
  • Offer agency: changing posture, opening eyes, stopping, or orienting to the room is always allowed.
  • Avoid commanding the listener to revisit traumatic memories.
  • For sleep audio, do not instruct a listener who may be driving or operating machinery to continue.

Timing model

Write for speech, not page reading.

  • Calm narration: roughly 90–120 spoken words per minute before pauses.
  • A 10-minute meditation usually needs about 650–900 spoken words depending on silence.
  • Mark silence deliberately: [pause 5s], [silence 20s].
  • Use longer pauses after sensory invitations and shorter pauses after orientation.

Do not fill every second with language. Silence is part of the composition.

Script architecture

  1. Arrival (10–15%) — context, agency, posture, external orientation.
  2. Settling (15–20%) — contact points, sound, ordinary breath.
  3. Core practice (35–45%) — one technique, progressively deepened.
  4. Integration (15–20%) — meaning without overclaiming; connect to the listener's real next moment.
  5. Return (10–15%) — wider sound, movement, eyes, choice, concise closing line.

For sleep, replace the active return with a permissive fade.

Language craft

  • Address one listener, not an audience.
  • Use short clauses and breathable sentence lengths.
  • Give one instruction at a time.
  • Ground imagery in sensation: temperature, weight, distance, texture, rhythm, sound.
  • Use a coherent image field; do not jump from ocean to stars to forest to fire in thirty seconds.
  • Alternate invitation and silence. Repetition should deepen, not pad.
  • Avoid guru language, cosmic certainty, moralized calm, and claims that the listener "should" feel a state.
  • End with a phrase simple enough to remember in daily life.

Core technique options

Choose one primary method:

  • external orienting and five-senses awareness;
  • contact-point body awareness;
  • natural breath observation;
  • gentle count with no retention;
  • open monitoring of sound/thought;
  • compassion or gratitude grounded in one real person/action;
  • future rehearsal of a specific ordinary moment;
  • sleep body-softening sequence.

Do not stack four techniques into one short meditation.

Companion Suno bed brief

Design the bed around narration:

  • instrumental, no vocals, no spoken samples;
  • sparse midrange so the voice remains intelligible;
  • no sudden transients, drops, risers, or dramatic key changes;
  • stable low-energy arc with subtle evolution every 30–60 seconds;
  • gentle pulse or free time according to purpose;
  • restrained low end and soft high frequencies;
  • long-tail ambience that does not mask consonants;
  • clean ending or extended fade.

Example shape (adapt, do not copy automatically):

Instrumental organic ambient, free slow pulse, felt piano harmonics, bowed glass, warm low drones and distant room air; spacious but speech-safe mix, restrained midrange, no percussion transients, no vocals, no choir, no dramatic swells; nearly still opening, subtle warmth at midpoint, long gentle fade.

Quality pass

Before delivery:

  • estimate spoken duration and show it;
  • verify opening includes agency and safe orientation;
  • verify no medical/clinical certainty;
  • verify breath language is optional and ordinary;
  • verify the core uses one technique;
  • verify pauses are explicit and sufficient;
  • read aloud for tongue-twisters, abstraction, and accidental urgency;
  • verify the music bed leaves space for speech.

Delivery format

Return:

  1. title and purpose;
  2. estimated narration duration;
  3. production lane;
  4. complete script with pause cues;
  5. separate instrumental Suno style prompt + exclusions;
  6. one safety note only if context calls for it.