Forbidden Configuration — Operationalizing Policy as Code to guard against...

OIDC client assertion. oidc_client_credentials now supports client assertion. API tokens are stored as digests. The database no longer keeps tokens in cleartext. Extensible source providers. Source...

A hands-on Kyverno policy-as-code guide: install it, write validate/mutate/generate policies, run audit vs enforce, and test policies in CI before they block prod.

Had the chance to attend An Evening with AI and Policy as Code hosted by AWS User Group Norway earlier this week. It was great to hear practical experiences around AWS Textract, Policy as Code with C...

Admission politikaları güvenlik onay kutusu değil, kümenin yazma yoluna eklenen senkron bağımlılıktır. Kyverno 1.19 sonrası üç eksende denge kurma rehberi.

Admission policies are not a security checkbox; they are a synchronous dependency on your cluster's write path. A guide to balance on three axes after Kyverno 1.19.

Why autonomous agents need hard limits, circuit breakers, and emergency stop mechanisms to prevent runaway execution and cascading failures.

Guardrails must be enforced at runtime, not left as developer best practices. Just like infrastructure-as-code, compliance must be baked into execution.

Organizations are increasingly using LLMs to generate policy-as-code (in languages like Rego and Cedar) to speed development, but AI-generated policies can be syntactically correct yet semantically flawed and grant unintended access. Research by Vatsal Gupta highlights recurring failure modes—missing contextual restraints, absent deny logic, hallucinated attributes, dropped temporal conditions, and action...