PPQS
BLOG

Notes from the input layer.

Data, analysis, and findings from scoring AI prompts across 7 verticals.

500 AI prompts scored across 8 quality dimensions. None passed.

I scored 500 software engineering prompts against an 8-dimension quality rubric. Every single one failed. Average score: 13.3/80. Zero graded C or above. Here's what that looks like and why it matters for the agent economy.

© 2026 OnChainIntel · The quality gate for prompts.