What problem does it solve?
This Skill solves the problem of turning small, unorganized sets of named personas into structured, domain-aligned judgment casts that respect explicit safety constraints, eliminating inconsistent or unsafe persona use in decision-making workflows.
Core Features & Use Cases
- Safety-Aligned Tiering: Assigns each persona to a tier defined by your caller-supplied domain safety rule, capping what each persona is allowed to emit to prevent overreach.
- Decision-Lens Mapping: Maps every persona to a concrete judgment perspective in your domain's vocabulary, with clear conditions for when to invoke each voice, and pairs adversarial personas with constructive counter-voices.
- Anchored Expansion: Proposes 2-4 additional personas that fill uncovered judgment lenses, with real sourced anchors for the strongest candidates to avoid shallow, unsubstantiated additions.
- Structured Output: Emits a ready-to-load personas.json artifact that integrates directly with your existing systems.
Use Case: For a software engineering team, input a seed set of "senior developer, QA engineer, product manager" plus a safety rule that prohibits personas from making final architectural decisions, and get a fully tiered, lens-mapped cast with additional relevant voices like "security auditor" anchored to real industry security standards.
Quick Start
Use the persona-roster-expander skill to turn your seed personas for your target domain into a tiered, judgment-mapped cast that strictly follows your specified safety rule.
