judgment-day

Delegate code reviews to Claude and Copilot CLI and synthesize verdicts.

Updated Apr 20, 2026

Search Tags:#memory #synthesis #ai-assistant #code-review #task-hub #dual-review

Ready to run

npx skills add https://github.com/juanmanueldaza/invencible --skill judgment-day-juanmanueldaza

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: judgment-day
Source: https://github.com/juanmanueldaza/invencible/tree/main/.claude/skills/judgment-day
Command: npx skills add https://github.com/juanmanueldaza/invencible --skill judgment-day-juanmanueldaza

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Independent blind reviews of code or changes to prevent bias and ensure quality, with synthesized verdicts used to drive decision-making.

Core Features & Use Cases

Phase 1: Blind delegation to Claude and Copilot CLI via the hub for parallel reviews.
Phase 2: Synchronous collection and synthesis of findings into a structured verdict table.
Phase 3: Iterative fix loop with escalation to humans when needed.

Quick Start

Invoke Judgment Day on a change set to run independent reviews and synthesize a final verdict.

Frequently Asked Questions about judgment-day

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

FAQPage Schema

What is a dual blind code review and how does it prevent bias?▼

A dual blind code review runs independent, parallel analyses from Claude and Copilot CLI to prevent shared bias. It synthesizes the separate findings into a structured verdict table, ensuring higher quality decision-making for code changes.

How do I run parallel code reviews using Claude and Copilot CLI?▼

You run parallel code reviews by invoking the process on a change set to delegate blind reviews to Claude and Copilot CLI. The system then synchronously collects the findings and synthesizes them into a final structured verdict.

Do I need a memory hub to coordinate dual code reviews?▼

Yes, you need a memory hub and hub delegation features to coordinate parallel review tasks and store the synthesized verdicts. The memory hub manages the independent delegations and collects the results for final aggregation.

What's the best way to synthesize independent code review findings?▼

The best way to synthesize independent code review findings is to use a structured verdict table that aggregates blind reviews from Claude and Copilot CLI. This approach combines parallel analyses into a unified decision-making output.

When do I need an iterative fix loop for code review verdicts?▼

You need an iterative fix loop during Phase 3 of the dual review process when synthesized verdicts require corrections. The loop handles continuous fixes and escalates issues to humans when the automated reviews cannot resolve them.

Can I escalate code review conflicts detected by parallel AI assistants?▼

Yes, you can escalate code review conflicts during the iterative fix loop phase. When the synthesized verdict table from Claude and Copilot CLI reveals unresolvable issues, the process escalates them directly to humans for final decision-making.