evidence-adjudicator

Synthesizes investigator evidence into ranked root cause verdicts with confidence scores.

8|1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/bordenet/superpowers-plus --skill evidence-adjudicator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: evidence-adjudicator
Source: https://github.com/bordenet/superpowers-plus/tree/main/skills/engineering/evidence-adjudicator
Command: npx skills add https://github.com/bordenet/superpowers-plus --skill evidence-adjudicator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill consolidates fractured investigator outputs after an incident to produce a single, evidence-weighted root cause verdict, resolving contradictions and avoiding over-counting agent agreement.

Core Features & Use Cases

  • Evidence synthesis: Aggregate supporting and disconfirming findings from multiple investigator branches and deduplicate repeated facts.
  • Reasoning trees & confidence: Build a structured reasoning tree with confidence-weighted scores and calibrated verdicts.
  • Adversarial validation: Run a disconfirmation pass to surface alternative explanations and reduce confirmation bias.
  • Use Case: After parallel investigator runs on a production outage, produce a ranked list of hypotheses with supporting evidence, gaps, and escalation flags for human review.

Quick Start

Ask the evidence-adjudicator to synthesize all investigator branch evidence and return a ranked root cause verdict with supporting and disconfirming evidence and confidence scores.

Frequently Asked Questions about evidence-adjudicator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I consolidate multiple investigator branches into a single root cause verdict?

Consolidate fractured investigator branches by synthesizing supporting and disconfirming evidence, deduplicating facts, and applying confidence scoring to produce a ranked root cause verdict.

What is the best way to resolve contradictions in post-incident investigation evidence?

Resolve contradictions in post-incident investigation evidence by applying contradiction detection and adversarial disconfirmation passes to surface alternative explanations and eliminate confirmation bias.

How does confidence scoring work for incident response root-cause analysis?

Confidence scoring for incident response root-cause analysis works by building a structured reasoning tree that weights supporting and disconfirming findings into calibrated verdicts.

Can I use evidence synthesis to avoid over-counting agent agreement during an outage review?

Yes, evidence synthesis deduplicates repeated facts across parallel investigator runs to avoid over-counting agent agreement and generate accurate escalation flags for human review.

How do I run an adversarial disconfirmation pass on incident reproduction results?

Run an adversarial disconfirmation pass on incident reproduction results by challenging the reasoning tree with alternative explanations, measuring divergence points, and adjusting confidence scores accordingly.

When should I not use automated evidence adjudication for production outage investigations?

Avoid automated evidence adjudication for production outages when investigator branches lack sufficient supporting or disconfirming evidence, as confidence-calibrated verdicts require structured branch inputs to function reliably.