skills.rest
PeterSalvato
PeterSalvatoCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

convergence

Analyze agreement and disagreement patterns across lens verdicts in audit results.

Updated Mar 4, 2026
Search Tags:#audit#evaluation#consensus#contradiction#meta-evaluation
One-click install
npx skills add https://github.com/PeterSalvato/formwork --skill convergence-petersalvato
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: convergence
Source: https://github.com/PeterSalvato/formwork/tree/main/skills/diagnostics/convergence
Command: npx skills add https://github.com/PeterSalvato/formwork --skill convergence-petersalvato

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill identifies where lenses agree and disagree in audit results, providing a meta-evaluation of the site's performance.

Core Features & Use Cases

  • Consensus Analysis: Maps consensus vs. contradiction across audit results.
  • Decision Points: Highlights contradictions where lenses disagree.
  • Invisible Pages: Identifies pages with no strong signal.
  • Tier Patterns: Detects systemic patterns in verdict clustering.
  • Verdict Map: Provides a comprehensive view of lens verdicts per page.

Quick Start

Run the convergence skill with the command: /convergence

Frequently Asked Questions about convergence

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

FAQPage Schema
How do I analyze audit results for consensus and contradiction across multiple lenses?▼

Meta-evaluation of audit results detects consensus and contradiction by comparing multiple lens verdicts. It identifies high-confidence agreement signals and highlights specific decision points where lenses conflict.

What is the best way to identify decision points when audit lenses disagree on verdicts?▼

Identifying decision points requires mapping contradictions where lenses disagree. Analyzing multiple lens verdicts highlights systemic patterns in verdict clustering and isolates pages lacking strong signals.

How do I detect systemic patterns in verdict clustering during a site audit?▼

Detecting systemic patterns in verdict clustering involves evaluating agreement and disagreement across lens outputs. Mapping tier patterns and comprehensive verdicts per page reveals systemic strengths and weaknesses.

Can I evaluate agreement and disagreement patterns without multiple lens outputs?▼

Evaluating agreement and disagreement patterns requires multiple lens verdicts and evidence. Without multiple lens outputs, mapping consensus, contradiction, and high-confidence signals is impossible.

How do I identify pages with no strong signal during a meta-evaluation?▼

Identifying pages with no strong signal involves finding invisible pages within audit results. These pages lack consensus or contradiction across lens verdicts, requiring further evidence for a confident decision.

When should I not use a convergence analysis for audit results?▼

Avoid convergence analysis when lacking multiple lens verdicts or evaluating a single isolated page. The process requires multiple lens outputs to effectively map consensus, contradictions, and tier patterns.

Related Skills

Explore More Software Engineering Skills→
hellonishhellonish
1

claim_verification

Map synthesis claims to cited sources and output per-claim verdicts.

Community
Advanced
ForgePlanForgePlan
3

refine

Refine design documents by resolving contradictions and updating CONTEXT.md and ADRs.

Official
Advanced
ognjhuntognjhunt

Blueprint Capture

Manage iOS capture app workflows and raw evidence bundle uploads.

Community
Intermediate
microsoftmicrosoft
10

council-here

Review the current conversation with a structured six-lens critique.

Official
Advanced