Santa Method Multi-agent adversarial verification framework. Make a list, check it twice. If it's naughty, fix it until it's nice. The core insight: a single agent reviewing its own output shares the same biases, knowledge gaps, and systematic errors that produced the output. Two independent reviewers with no shared context break this failure mode. When to Activate Invoke this skill when: - Output will be published, deployed, or consumed by end users - Compliance, regulatory, or brand constraints must be enforced - Code ships to production without human review - Content accuracy matters (tech…