ratchet:score

Calculate debate pair quality metrics and trends from repository artifacts.

1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/netbrain/ratchet --skill ratchet-score
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ratchet:score
Source: https://github.com/netbrain/ratchet/tree/main/skills/score
Command: npx skills add https://github.com/netbrain/ratchet --skill ratchet-score

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Consolidates debate and review artifacts to surface per-pair quality metrics, surface regressions, and provide an evidence-backed view of agent pair performance so teams can prioritize tuning and interventions.

Core Features & Use Cases

  • Aggregate debate metadata: Read .ratchet/debates/*/meta.json to compute total debates, consensus rates, rounds, verdict breakdowns, and fast-path usage per pair.
  • Ingest review signals: Read .ratchet/reviews/<pair>/review-*.json to compute average effectiveness, gen vs adv splits, and top missed issues.
  • Persist and smooth history: Update .ratchet/scores.yaml using an exponential moving average (α=0.3) so pair metrics survive archival and support trend analysis.
  • Operational use cases: Track pair health over time, decide when to tighten agents, drill down into recent debates, and guide next debate runs.

Quick Start

Run /ratchet:score to display quality metrics and trends for the current workspace or provide a pair name to focus on a single pair.

Frequently Asked Questions about ratchet:score

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

FAQPage Schema
How do I track code review quality metrics for agent pairs over time?

Track code review quality metrics by calculating consensus rates, average rounds, and verdict distributions from debate artifacts, then persisting them with an exponential moving average to smooth historical trends.

What is the best way to measure debate consensus rates and review effectiveness?

Measure debate consensus rates and review effectiveness by ingesting per-pair metadata and review JSON artifacts to compute average effectiveness, gen vs adv splits, and top missed issues.

How do I calculate quality scores from archived debate and review artifacts?

Calculate quality scores from archived artifacts by reading workspace metadata and review JSON files, computing epoch metrics, and using historical fallbacks to maintain trend accuracy.

Does the exponential moving average smoothing work for cross-workspace pair comparisons?

Yes, exponential moving average smoothing works for cross-workspace comparisons by applying an alpha of 0.3 to update scores, allowing pair metrics to survive archival and support trend analysis.

Why do I need to monitor fast-path usage in agent pair debates?

Monitor fast-path usage in agent pair debates to evaluate review effectiveness and identify regressions, providing an evidence-backed view of agent performance to prioritize tuning and interventions.

Can I analyze a single agent pair's performance instead of the whole workspace?

Yes, you can analyze a single agent pair's performance by providing the specific pair name, which focuses the quality metrics and trend analysis on that pair alone.