fragility-measurement

---
name: fragility-measurement
description: Compute a fragility index from flip-point distances and degradation scores, summarizing how robust the conclusion is.
execution: subagent
prompt: ./prompt.md
input: flip_points (list), degradation_scores (list), factor_count (number)
used-by: [counterfactual-probing]
---

# Fragility Measurement

Aggregates flip-point distances and degradation scores into a single fragility index.

## Execution

Subagent — spawned via subagent-spawning/spawn-agent.

## Why Subagent

Fragility computation requires synthesizing results from multiple ablation and flip-point tests into a coherent assessment.

## Input

- **flip_points**: List of {dimension, distance, confidence}
- **degradation_scores**: List of {factor, score, classification}
- **factor_count**: Total factors examined

## Output

- **fragility_index**: Overall fragility (0.0 = robust, 1.0 = extremely fragile)
- **most_fragile_dimension**: Where the conclusion is most vulnerable
- **robustness_assessment**: Qualitative summary
- **recommendations**: What to investigate further

## Budget

One unit = one aggregation per campaign run.