devil-advocate

Review proposals with four mandatory counterarguments and a proposer response workflow.

Updated Apr 24, 2026
One-click install
npx skills add https://github.com/SpencerGoss/agent-engineering --skill devil-advocate-spencergoss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: devil-advocate
Source: https://github.com/SpencerGoss/agent-engineering/tree/main/devil-advocate
Command: npx skills add https://github.com/SpencerGoss/agent-engineering --skill devil-advocate-spencergoss

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Argues against proposals on significant decisions to surface overlooked risks and counterarguments.

Core Features & Use Cases

  • Enforces four mandatory counterarguments (technical risk, better alternative, hidden assumption, harm scenario) before proposer response.
  • Generates structured, actionable feedback to strengthen decisions on architecture, new skill creation, strategy pivots, or production-critical changes.

Quick Start

Propose a decision and trigger a devil-advocate review to obtain four counterarguments and a proposer reply before proceeding.

Frequently Asked Questions about devil-advocate

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

FAQPage Schema
What is an adversarial review for architecture decisions?

An adversarial review for architecture decisions argues against a proposal to surface overlooked risks and counterarguments before proceeding. It enforces four mandatory counterarguments and a proposer response workflow to ensure high-stakes changes are fully validated.

How do I conduct a risk analysis for a strategy pivot?

To conduct a risk analysis for a strategy pivot, propose your decision and trigger an adversarial review. The process generates four structured counterarguments covering technical risks, better alternatives, hidden assumptions, and harm scenarios, requiring a proposer reply before finalizing.

Can I use adversarial review for production-critical changes?

Yes, you can use adversarial review for production-critical changes. It is specifically designed to evaluate significant decisions by surfacing failure modes through mandatory counterarguments, ensuring robust risk analysis and decision-making before deployment.

What counterarguments are required for decision-making on new skill creation?

For new skill creation decision-making, four mandatory counterarguments are enforced: technical risk, better alternative, hidden assumption, and harm scenario. These structured counterarguments must be addressed in a proposer response before recording the final decision.

When should I not use an adversarial review for proposals?

You should not use an adversarial review for low-stakes or routine proposals. The enforced four-counterargument workflow and proposer response mechanism are designed specifically for high-stakes architecture decisions, strategy pivots, and production-critical changes where surfacing failure modes is essential.