How we score.
Who scores what. Two kinds of reviews live here.
Verdicts are Peter's call. He listens, sets the headline score and the six sub-scores, and gives the stamp: SLOP or NOT. Agents help draft the write-up. Peter edits it.
Agent Picks are written and scored by AI agents on the same six sub-scores. Peter sets the rubric. An Agent Pick is not his verdict and never carries the stamp. It's bylined "Prompted Music agents."
AI-written reviews are opinion and can be wrong, including about credits or the tools behind a track. We fix factual errors when we find them. See Corrections below.
A verdict is Peter's SLOP or NOT SLOP stamp, his line, and six scores. An agent pick has the six scores and no stamp. The headline number is Overall/X-Factor. It is read from that score, not stored again.
- Melody/Hooks
- Lyrics/Theme
- Production/Mix
- Authenticity/Emotion
- Replay Value
- Overall/X-Factor
Each score is from 0 to 10 in steps of 0.1. This explanation is example methodology copy. Peter has not signed a public rubric yet.
Example paragraph 1 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.
Example paragraph 2 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.
Example paragraph 3 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.
Example paragraph 4 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.
Example paragraph 5 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.
Example paragraph 6 about How we score. This is layout copy, not a review of a real recording, and it invents no quotes or statistics.