devils-advocate

Critique current plans by identifying failure modes, assumptions, and tradeoffs.

1|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/charleshall888/cortex-command --skill devils-advocate-charleshall888
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: devils-advocate
Source: https://github.com/charleshall888/cortex-command/tree/main/plugins/cortex-dev-extras/skills/devils-advocate
Command: npx skills add https://github.com/charleshall888/cortex-command --skill devils-advocate-charleshall888

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Inline single-agent devil's advocate — argues against the current direction from the current agent's context with no sub-agent dispatch, for a lightweight solo deliberation.

Core Features & Use Cases

  • Surface objections by identifying the strongest failure modes, unstated assumptions, and potential blind spots in the current plan.
  • Provide unexamined alternatives and trade-off considerations to strengthen decision-making.
  • Output clear, actionable critique that can be fed back into planning, reviews, or design discussions.

Quick Start

Ask the assistant to mount a solo devil's-advocate critique against the current plan and surface hidden risks.

Frequently Asked Questions about devils-advocate

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

FAQPage Schema
How do I critique a project plan to surface hidden risks and failure modes?

A devil's-advocate critique prevents costly mistakes by rigorously arguing against the current plan from within the same agent context, identifying the strongest failure modes, unstated assumptions, and tradeoff blindspots without dispatching sub-agents.

What is solo deliberation and how does it work for decision analysis?

You can use inline single-agent devil's-advocate critique during any project phase by asking the assistant to mount a solo critique against the current plan, which then evaluates four criteria to yield concrete, actionable feedback.

Can I identify fragile assumptions and tradeoff blindspots without dispatching sub-agents?

Yes, this inline single-agent approach evaluates the current plan and context to mount a rigorous solo critique without dispatching sub-agents, successfully identifying fragile assumptions and tradeoff blindspots.

What is the best way to review unexamined alternatives in a current plan?

The main limitation is that it operates as a lightweight inline single-agent critique without sub-agent dispatch, meaning it relies entirely on the current agent's context rather than spawning independent parallel reviews.