---
name: yogsoth-ai/knowledge-structuring-novelty-scoring
source: https://app.decimal.ai/s/yogsoth-ai-knowledge-structuring-novelty-scoring@1/SKILL.md
source_sha256: a4af969201de
---

# Novelty Scoring

Score empty cells (unexplored combinations) by their novelty potential.

## Tool

`vault_search` + CC reasoning

## Protocol

1. For each empty cell, assess:
   - Distance from nearest occupied cell (more distant = more novel)
   - Feasibility (are there physical/logical constraints?)
   - Potential impact (would this combination solve known problems?)
2. Score 0-10: novelty (0=obvious, 10=paradigm-shifting)
3. Flag top-scoring cells as priority research opportunities

## HARD-GATE

<HARD-GATE>
Must provide explicit reasoning for each score, not just a number.
</HARD-GATE>

## Yield

Returns: `{ scored: Array<{combination: string, score: number, reasoning: string}> }`