Wargame 5 user archetypes with romancer4 behavioral simulation

[jw409]

Vision (Updated 2026-03-31)

Use romancer4's multi-agent behavioral simulation to model how 5 different user archetypes interact with kinderpowers.

New direction: This is a public repo. Rather than coupling to romancer4 directly, kinderpowers defines a simulation client extensibility point — a structured JSON interface for scenarios and results that any simulation engine can implement.

What Changed

• Simulation-specific logic moves to the romancer4 side
• Kinderpowers provides: scenario format, result format, 5 standard archetypes, integration patterns (skill, agent, MCP)
• Any tool that reads scenarios and writes results works — romancer4, custom scripts, other behavioral simulation tools
• See docs/simulation-extensibility.md for the full spec

The 5 Standard Archetypes

1. The Beginner — barely codes, just got Claude Code, doesn't know git. Success = "Claude is helping me learn."
2. The Senior Dev — knows what they want, skeptical of AI frameworks. Success = "This saves me time without overhead."
3. The Team Lead — needs to scale across a team. Success = "My team ships faster."
4. The Open Source Maintainer — needs to triage and review. Success = "I process PRs in half the time."
5. The AI-Native Builder — builds on top of AI tools. Success = "This gives me capabilities I couldn't build alone."

Remaining Work

• Create var/archetypes/ with JSON files for each archetype
• Create a wargame skill that orchestrates scenario execution
• Wire romancer4 as a reference simulation client implementation
• Run first simulation pass against current kinderpowers and document results