Music brief: describe the track you need before you search for it

A fill-in brief that names mood as harmony, a BPM range, genre, length and where the drop must land relative to your hook, so a library search or a composer returns something that fits the cut.

MUSIC BRIEF for {video title}
LENGTH: {total seconds} s; the track may be longer, the edit will cut it.
MOOD, as harmony, not adjectives: {e.g. 'major sevenths and a minor sixth, solitary and warm' / 'triads with an added sixth, no minor chords, light'}
TEMPO: {low BPM}-{high BPM} BPM. The picture is cut at about one cut per {seconds per cut} s, so a bar should be close to that.
GENRE / INSTRUMENTS: {two or three instruments and one texture, e.g. 'felt piano, upright bass, brushed drums, tape hiss'}
STRUCTURE RELATIVE TO THE PICTURE:
- 0 to {hook end} s: under the hook, sparse, no drums; it must not compete with the first line.
- Drop, or the first full-band bar, at exactly {drop time} s: the cut right after the hook.
- A thinner section from {breakdown start} s to {breakdown end} s, where people talk.
- End: a last chord that rings out, not a fade.
VOCALS: none. No lyrics, no vocal chops.
AVOID: {what fights the footage, e.g. 'no risers, no trailer hits'}
SPACE FOR HOLES: the score will be silenced from {hole start} s to {hole end} s where live sound plays; write nothing that needs those bars.
DELIVERABLE: WAV, 48 kHz, plus a stem of the drums alone so the edit can mute them under speech.

Paste into your assistant as the first message of a new chat; fill the braces first.

Steps

  1. Fill the brief from the finished cut, not before it: the drop time is a fact about your edit, not a wish.
  2. Paste it into a library's search as keywords (mood words, BPM, instruments) or send it as is to a composer or a music generation tool.
  3. Audition the candidate against the picture with the voice track on. A track that sounds right alone and wrong under speech is the usual outcome.

Tips

  • Mood is scale degrees, not tempo: a 'major' score came out sad at the right BPM because of its chords. Say the chords, not the feeling.
  • Mix the score about 15 dB under the voice for talk films. A brief that asks for 'subtle' gets a quiet track that still fights the voice in the same range; ask for instruments that sit away from speech.
  • If someone plays in the footage, the key and chords can be lifted from the recording with audio-to-MIDI and handed to the composer; the melody usually cannot.

Read more

works withClaude: Paste into claude.ai or the Claude app. · ChatGPT: Paste into ChatGPT; works with any GPT model. · Any LLM: Model-agnostic: Gemini, Mistral, a local model. · By hand: No tool needed: a checklist or a technique.
costFree. No API, no key.
statusUnverified: written from the technique, not yet run end to end by us.
licenseCC BY 4.0 for the text here; skills are MIT.
updated2026-11-01
for agentsmusic-brief-prompt-for-picking-a-track.md
#music-brief#score#track-selection#bpm#mood#prompt-template

Related

Music & sound beginner

Prompt for a music generator: a 30-second instrumental for a short

A brief in the shape music generation tools read: structure by seconds, BPM, three instruments, no vocals, a drop that lands on your hook cut, and an ending that stops instead of fading.

Any LLMClaudeChatGPT
Music & sound intermediate

Beat-sync cutting: cut on the beat, not on every beat

A rule and an agent prompt for cutting to music from a list of beat timestamps: cuts land on beats, phrases get room to breathe, and the best shot is held through the hook.

Claude CodeClaudeChatGPT
Music & sound advanced

Mashups and ciphers with a beat grid: sdsp-mash

Pick pairs by BPM and key, split stems, stretch only the beat, and render one beat with verses from other songs, or a beat switch per verse, from a TOML file.

Claude Code
Music & sound intermediate

Score with a hole, not ducking, where something plays live

Cut the score to silence under live music, a laugh or a knock instead of ducking it; ducking is audible, a hole is not. The rule, the measured levels and the ffmpeg line.

ffmpegCapCut / any editorClaude Code
Music & sound intermediate

Recording to MIDI: pick the model by material, then check it

Three local transcription models fail in different directions. The skill picks one by what the material is, runs it, and checks the result before you trust it.

Claude Code
Skills & tools intermediate

Beat times and downbeats from a song in Node, with life2film-engine

A short Node script that decodes a song with ffmpeg and prints BPM, beat times and downbeats on your machine, ready to place cuts on. No upload, no API key.

Claude CodeffmpegLife2Film