# Impact Scoring

> SOP: assess the potential impact of a research gap, identify beneficiaries and output an impact score

- Skill: `yogsoth-ai/impact-scoring` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add yogsoth-ai/impact-scoring`
- Raw SKILL.md: https://api.skillmd.com/api/skills/yogsoth-ai/impact-scoring/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: yogsoth-ai (https://skillmd.com/u/yogsoth-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/yogsoth-ai/impact-scoring

---


# Impact Scoring

Assess the potential impact of a research gap, identify beneficiaries and output an impact score.

## HARD-GATE

<HARD-GATE>
- Input must be a GapRecord with status: "complete"
- The output composite score must lie in the [1, 5] range
- The beneficiaries list must not be empty (at least 1 beneficiary)
- Each sub-dimension must carry at least 1 sentence of textual rationale
</HARD-GATE>

## Pipeline

1. **Precondition check**: verify input GapRecord completeness
2. **Beneficiary identification**: enumerate the direct and indirect beneficiaries once this gap is filled (researchers, practitioners, patients, policymakers, society, etc.)
3. **Impact breadth assessment**: the breadth of beneficiaries — the size of the affected population/institutions (1-5)
4. **Impact depth assessment**: the degree of change for beneficiaries — marginal improvement or fundamental transformation (1-5)
5. **Composite score**: equal-weighted average of the two dimensions, to one decimal place
6. **Output**: return the ImpactScore object

## Output Format

```json
{
  "gap_id": "gap_001",
  "beneficiaries": [
    { "group": "beneficiary name", "type": "direct | indirect", "description": "how they benefit" }
  ],
  "dimension_scores": {
    "breadth": { "score": 4, "rationale": "..." },
    "depth": { "score": 3, "rationale": "..." }
  },
  "composite_score": 3.5,
  "overall_rationale": "Overall rationale (2-4 sentences)"
}
```