skeptical-review

Evaluate proposed ideas and plans against evidence and constraints.

2|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/bdemarzo/ai-workflows --skill skeptical-review-bdemarzo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skeptical-review
Source: https://github.com/bdemarzo/ai-workflows/tree/main/skills/skeptical-review
Command: npx skills add https://github.com/bdemarzo/ai-workflows --skill skeptical-review-bdemarzo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents avoidable downstream rework by forcing earlier, explicit scrutiny of the strongest claims in an idea, spec, plan, or implementation summary before moving forward.

Core Features & Use Cases

  • Artifact-focused critique: Challenges the exact decision, artifact, or claim being proposed rather than drifting into generic feedback.
  • Consequence-driven skepticism: Prioritizes fatal flaws, likely failure modes, missing evidence, and operational/delivery risks over minor nits.
  • Evidence-based accountability: Explicitly states what evidence or changes would reverse the conclusion, and separates fatal issues from optional improvements.
  • Stays optional and non-owning: Works as a manual pressure-test and does not replace the required internal reviewer inside formal workflow phases.

Quick Start

Ask the AI to skeptical-review your current artifact or decision and return a severity-grouped critique with a single recommendation.

Frequently Asked Questions about skeptical-review

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

FAQPage Schema
How do I pressure-test product decisions before committing to implementation?

Spec validation evaluates whether proposed plans are justified by evidence and constraints, grouping findings by severity to isolate fatal flaws from optional improvements before you commit to implementation.

What is a skeptical review and when do I need it for architecture critique?

A skeptical review is an optional manual pressure-test for architecture choices that identifies missing evidence and operational risks, needed when you want explicit scrutiny outside formal workflow review stages.

How do I assess delivery constraints and failure modes in a project plan?

Assess delivery constraints by applying consequence-driven skepticism to the plan, prioritizing likely failure modes and fatal risks while explicitly stating what evidence would reverse the conclusion.

Can I use this approach to challenge an implementation summary without replacing formal review?

Yes, you can challenge an implementation summary as a manual pressure-test that stays optional and non-owning, preserving the source artifact while providing critique without replacing required internal reviewers.

What are the limitations of using a skeptical review for risk assessment?

The main limitation is it avoids file creation and does not own the workflow; it serves only as an optional manual pressure-test, meaning it cannot replace formal review stages or enforce required architectural changes.