skills.rest
alatyshau
alatyshauCOMMUNITY
Explore all skills by this author →
GitHub

critical

Reviews code and designs to find problems, edge cases, and security issues.

Updated Nov 21, 2025
Search Tags:#code review#edge cases#security audit#critical thinking#design validation#bug finding
One-click install
npx skills add https://github.com/alatyshau/duet --skill critical-alatyshau
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: critical
Source: https://github.com/alatyshau/duet/tree/main/packages/skills/stances/critical
Command: npx skills add https://github.com/alatyshau/duet --skill critical-alatyshau

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reviews often default to polite confirmation instead of genuine scrutiny, letting logic errors, security holes, and edge cases slip through. This Skill forces an adversarial review stance that actively hunts for what is wrong rather than confirming what is right. ## Core Features & Use Cases - Adversarial Review Stance: Assumes problems exist and systematically searches for logic errors, security holes, missing validation, and unclear naming. - Edge Case Coverage: Checks empty input, null values, maximum values, and concurrent access scenarios. - Severity-Prioritized Output: Structures findings as issue, impact, and fix, grouped by severity with a summary count. - Use Case: Before merging a pull request, ask for a critical review and receive a prioritized list of concrete issues with fixes instead of a vague approval. ## Quick Start Ask the AI to critically review your code or design and list every problem it can find with a fix for each.

Frequently Asked Questions about critical

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

FAQPage Schema
How do I get an AI to critically review my code?▼

Ask for a critical review that assumes problems exist and hunts for them. This stance checks edge cases like empty input, null values, and concurrent access, then reports each issue with its severity and a concrete fix.

What should a good code review checklist include?▼

A thorough review covers logic errors, security holes, missing input validation, unclear naming, and edge cases such as empty input, null, maximum values, and concurrent access. Findings should be prioritized by severity with fixes attached.

When should I use a critical review stance instead of a normal review?▼

Use it for code reviews, security audits, design validation, and testing edge cases, or whenever you ask what could go wrong. Switch to a pragmatic implementation stance once the review is complete and fixes should be applied.

Why do AI code reviews often miss obvious problems?▼

Default assistant behavior tends toward agreeableness, softening critique and leading with praise. An explicit critical stance removes that bias by requiring problems to lead the output and forbidding vague suggestions without fixes.

What are the limitations of an adversarial review stance?▼

It is not suited for implementation work or morale-sensitive feedback contexts. Once findings are delivered and the user asks to apply fixes, the stance should switch to pragmatic execution rather than continued fault-finding.

Related Skills

Explore More Software Engineering Skills→
ngochuy13ngochuy13

debug

N/A

Community
Intermediate
Moonsong-LabsMoonsong-Labs
3

requesting-code-review

Dispatch the code-reviewer subagent to review code and categorize findings by severity.

Official
Intermediate
eliottbusinesseliottbusiness

requesting-code-review

Automate pre-commit code review with static analysis and security scanning.

Community
Advanced
hung-phanhung-phan

property-based-testing

Generates and validates code behavior across multiple scenarios using property-based testing.

Community
Advanced