What problem does it solve?
This Skill eliminates the manual, inconsistent work of coordinating multiple review perspectives (external user, adversarial, expert) for artifacts like skills, code, and design documents. It automates target profiling, parallel persona dispatch, finding triage, and fix commitment, ensuring thorough, consistent validation without hours of manual coordination.
Core Features & Use Cases
- Automated Target Profiling: Classifies artifacts on 5 dimensions (artifact type, audience, claim density, risk level, novelty) to recommend the optimal simulation areas and persona composition.
- Parallel Multi-Persona Simulation: Dispatches beginner, expert, challenger, and custom personas as parallel agents to catch onboarding friction, claim-evidence gaps, security vulnerabilities, and edge cases that single-perspective reviews miss.
- Risk-Aware Gating: Enforces cross-model coverage requirements for medium and high-risk targets to avoid homogeneous blind spots, and withdraws auto-commit privilege for unvalidated same-session panels.
- Use Case: Before publishing a new Claude Code skill to a public marketplace, run sim-conductor to automatically run external user, adversarial, and expert reviews, get M/S/R triaged findings, and commit mandatory fixes in minutes instead of hours of manual review.
Quick Start
Invoke the sim-conductor skill with your target artifact path and desired simulation area to automatically profile the target, dispatch parallel review personas, triage findings, and commit mandatory fixes.
