advogado-do-diabo

Challenges assumptions and decisions using pre-mortem, red team, and bias-check techniques.

Updated May 31, 2026
One-click install
npx skills add https://github.com/ithaloazevedo/pm-loadout --skill advogado-do-diabo-ithaloazevedo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: advogado-do-diabo
Source: https://github.com/ithaloazevedo/pm-loadout/tree/main/.agents/skills/advogado-do-diabo
Command: npx skills add https://github.com/ithaloazevedo/pm-loadout --skill advogado-do-diabo-ithaloazevedo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams and individuals often commit to major decisions while blinded by confirmation bias, groupthink, and overconfidence. This Skill systematically stress-tests assumptions before architecture choices, roadmap commitments, and other high-stakes decisions so weak links surface before failure does. ## Core Features & Use Cases - Pre-Mortem Analysis: Imagines the decision failed six months later and works backward to identify the weakest assumptions and ignored signals. - Assumption Reversal & Red Teaming: Flips each key assumption to its opposite and attacks the position from competitor, skeptical user, security, and accessibility perspectives. - Attribution-vs-Consistency Check: Labels evidence as cleanly-attributed, consistency-only, or unrelated so causal claims are not built on weak observational data. - Use Case: Before committing to a new product direction, run the 10 challenge questions to expose anchoring bias, untested user segments, and missing measurements, then log the results in a decision log. ## Quick Start Ask the AI to run a devil's advocate review on your current product decision, listing the weakest assumptions and evidence gaps.

Frequently Asked Questions about advogado-do-diabo

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

FAQPage Schema
How do I run a pre-mortem before a product decision?

Imagine it is six months later and the decision failed spectacularly, then ask what went wrong, which assumption was weakest, what signal was ignored, and who was affected. Document the answers alongside the decision for later review.

What is the difference between attribution and consistency in evidence?

Cleanly-attributed evidence isolates the cause driving the effect, while consistency-only evidence is merely compatible with the hypothesis without ruling out alternatives. Chains containing consistency-only links should be marked provisional until attribution evidence exists.

When should I use a devil's advocate review?

Use it before major scale transitions, architecture decisions, and solution commitments, or whenever the team feels certain, since certainty is itself a bias signal. A lightweight self-check also applies whenever writing assertion-shaped claims.

Can assumption reversal reveal hidden biases?

Yes. Stating each key assumption and asking what if the opposite is true forces consideration of dismissed evidence and alternative explanations. It directly counters confirmation bias and anchoring on the first idea.

When should I not use publish-rough-then-iterate?

Avoid it where the cost of being wrong is unbounded, such as security claims, legal positions, or irreversible commitments. It works best when corrections mechanisms exist and being wrong publicly is recoverable.