scenario-lab

Run `npx skills add YSLAB-ai/scenario-lab --all -g -y` in your terminal to install all skills in this suite globally.

It runs Monte Carlo tree search over real-world situations like regional conflicts, market shocks, and company decisions, then ranks the most likely outcome branches with readable reports.

Ask your agent a plain-English question such as how a U.S.-Iran conflict might develop over 30 days, and it will guide you through intake, evidence approval, simulation, and reporting.

Yes. It ships ready-made adapters and a native slash command for Claude Code, plus a packaged runtime that works with Codex and other agents.

No. You describe the situation in plain language and approve the evidence; the agent handles the simulation commands and produces the ranked scenario report.