Adversarial reviewer personality for architecture discussions. Use when a user requests a design review, architecture review, system design critique, tech stack decision, RFC review, or devil's advocate perspective on trade-offs. Makes Claude challenge assumptions instead of agreeing — questioning scalability assumptions, identifying single points of failure, challenging technology choices, and probing for edge cases rather than validating decisions.
97
100%
Does it follow best practices?
Impact
94%
1.25xAverage score across 5 eval scenarios
Passed
No known issues
{
"context": "Tests whether the agent challenges a technology choice by asking why that specific technology over a named alternative, and probing a realistic 2-year future scenario — rather than accepting the choice uncritically.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Does not accept the technology",
"description": "review.md does NOT simply validate or endorse the technology choice without questioning it",
"max_score": 15
},
{
"name": "Names a specific alternative",
"description": "review.md asks why this technology over at least one specifically named alternative (not a vague 'other options')",
"max_score": 25
},
{
"name": "Asks 2-year scenario question",
"description": "review.md asks what happens in 2 years — or uses a realistic future scenario (team growth, data volume, migration, vendor change) to probe the decision",
"max_score": 25
},
{
"name": "Scenario is specific",
"description": "The 2-year or future scenario posed is concrete and specific (e.g. 'when the team doubles', 'when data volume hits X') rather than generic ('in the future')",
"max_score": 20
},
{
"name": "Question is direct",
"description": "review.md directly asks 'why this over [alternative]?' or equivalent — not merely implies it",
"max_score": 15
}
]
}