shadow-reviewer

Identify latent issues in PRs across the Sentinel project.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/dsieczko/claude-agent-plugins --skill shadow-reviewer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: shadow-reviewer
Source: https://github.com/dsieczko/claude-agent-plugins/tree/main/shadow-reviewer-agent/skills/shadow-reviewer
Command: npx skills add https://github.com/dsieczko/claude-agent-plugins --skill shadow-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Shadow Reviewer identifies latent issues in PRs, such as silent failure modes, test gaps, edge cases, and premature abstractions. It provides a structured adversarial perspective to surface risks before changes merge.

Core Features & Use Cases

  • Adversarial PR review focused on structural integrity, observability, and documentation alignment.
  • Spot silent failures, missing tests, and edge cases that consensus often overlooks.
  • Route concerns to CTO/Architect when escalation is needed to resolve high-risk issues.

Quick Start

Explain to Shadow Reviewer the PR context and request a thorough adversarial review focused on latent failures and missing coverage.

Frequently Asked Questions about shadow-reviewer

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

FAQPage Schema
How do I find silent failure modes and edge cases in a pull request?

To find silent failure modes during PR review, provide the code context to an adversarial reviewer to surface latent issues like missing error handling, test gaps, and premature abstractions before merging.

What is adversarial code review and when do I need it for risk management?

Adversarial code review applies a structured opposing perspective to identify structural integrity issues, observability gaps, and documentation misalignment. You need it when consensus review overlooks high-risk latent failures in architectural decisions.

How do I perform an independent second review on a pull request for observability gaps?

Perform an independent second review by routing the PR context to a shadow reviewer that evaluates data handling, error handling, and observability independently, yielding actionable suggestions and explicit risk signaling.

Can I escalate high-risk architectural issues identified during a PR review?

Yes, you can escalate high-risk architectural issues identified during PR review by routing concerns directly to a CTO or Architect, ensuring critical problems like silent failures and missing test coverage are resolved before merging.

Does adversarial PR review work for finding missing test coverage and premature abstractions?

Adversarial PR review effectively identifies missing test coverage and premature abstractions by challenging consensus, analyzing error handling, and evaluating data handling to provide actionable suggestions for code quality improvement.