The following pages link to detecting-ai-model-prompt-injection-attacks skill (Anthropic-Cybersecurity-Skills):