Blog
Notas sobre seguridad de agentes de IA con políticas deterministas: manifiestos firmados, enforcement y auditoría a prueba de alteraciones.
Más en inglés
3 min de lecturaen
A model can be persuaded. A signed policy cannot.
Prompt rules are not a security boundary for AI agents. Four ideas on moving the decision outside the model — and a rule you can write today.
securitypolicyprompt-injection
2 min de lecturaen
What a hash-chained audit log actually proves
Tamper-evident is not tamper-proof. Four ideas on what a chained log guarantees, what it does not, and how to check your own.
auditintegrityspec