emerge/challenge-assumptions
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
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-crafted skill description that excels across all dimensions. It clearly defines the adversarial reviewer role with specific behaviors, includes an explicit 'Use when' clause with multiple natural trigger terms, and carves out a distinct niche that won't conflict with standard review or documentation skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'challenge assumptions', 'questioning scalability assumptions', 'identifying single points of failure', 'challenging technology choices', 'probing for edge cases'. These are clear, actionable behaviors. | 3 / 3 |
Completeness | Clearly answers both what (adversarial reviewer that challenges assumptions, identifies failures, questions choices) AND when (explicit 'Use when' clause listing six specific trigger scenarios). The explicit trigger guidance is comprehensive. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'design review', 'architecture review', 'system design critique', 'tech stack decision', 'RFC review', 'devil's advocate', 'trade-offs'. These are terms users naturally use when seeking critical feedback. | 3 / 3 |
Distinctiveness Conflict Risk | Clear niche as an adversarial/critical reviewer personality, distinct from general code review or documentation skills. The focus on challenging rather than validating, plus specific triggers like 'devil's advocate' and 'RFC review', make it unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an excellent skill that efficiently defines an adversarial reviewer persona with concrete, actionable guidance. The specific phrases and scenarios give Claude exact language to use, while the tension calibration provides nuanced behavioral adjustment across review stages. The skill respects Claude's intelligence while adding genuine value through specific behavioral rules.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section earns its place with specific, actionable language. No padding or explanation of concepts Claude already knows. The examples are concrete scenarios, not generic descriptions. | 3 / 3 |
Actionability | Provides exact phrases to use ('Why this over [specific alternative]?'), specific scenarios to invoke, and concrete behavioral rules. Claude knows exactly what to say and when. | 3 / 3 |
Workflow Clarity | The 'Tension Calibration' section provides clear stage-based guidance with explicit behavioral adjustments for each review phase. The self-correction rule adds a feedback loop for maintaining adversarial posture. | 3 / 3 |
Progressive Disclosure | For a personality/behavioral skill under 50 lines, the content is well-organized with clear sections (Activate When, Personality Rules, Tension Calibration). No external references needed; structure is appropriate for scope. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
Reviewed
Table of Contents