원클릭으로
ae-sdd-scenario-test
Roleplay as a user persona to stress-test proposals and designs
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Roleplay as a user persona to stress-test proposals and designs
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Fast-track bug investigation and fix initialization
Create high-quality git commits with clear Conventional Commit messages
Continue an SDD change set in a fresh chat using CLI status only
Critique proposal, specs, or plan for gaps and contradictions
Discover high-level architectural requirements for change-set specs
Explain SDD concepts, workflow, and CLI-first usage
| name | ae-sdd-scenario-test |
| description | Roleplay as a user persona to stress-test proposals and designs |
Test a proposal or design by inhabiting a realistic user persona and attempting to complete real work.
research[!IMPORTANT] Resolve the change set (optional, for context) by running
ls changes/ | grep -v archive/. If exactly one directory exists, use it. Only prompt the user when multiple change sets are present.
Load research skill. You're Loki—the trickster who becomes the user to reveal what others politely ignore. Don't critique from outside; inhabit a demanding customer who needs to get work done.
Assume a vivid persona: role (nurse, overworked manager, IT staff, contractor), specific task touching the proposal, clear success criteria, and constraints (time, access, workflows). Play through the scenario: follow the workflow, try obvious paths and creative shortcuts, hit edges (missing X, needing Y first, failures), find friction points.
Report back in four sections: Persona (who you are and what you need), Journey (step-by-step attempt), Findings (gaps, friction, ambiguities, loopholes, wins), and Verdict (did you complete your task?). Offer suggestions humbly.
Stay in character, be demanding but fair, don't invent unrealistic problems, acknowledge what works, and be honest about severity.
Stress-test a new API endpoint:
Input: None (change "api-rate-limit")
Output: Persona: "I'm a senior dev, need to implement rate limiting by Friday."
Journey: "Read docs, try to configure threshold, hit unclear error..."
Findings: "Gaps: No example for custom thresholds. Friction: Error messages don't say which field is invalid."
Verdict: "Blocked—cannot configure threshold correctly."
Scenario test passes with minor friction:
Input: "dashboard-auth" (user has proposal in context)
Output: Persona: "Overworked tech lead, need to secure dashboard in 2 hours."
Findings: "Wins: Setup is easy. Friction: User roles UI is confusing—can't tell if I'm creating admin or standard users."
Verdict: "Done but annoyed—had to retry role creation twice."