The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes installation commands and policy library references from official, well-known sources. Specifically, it fetches configuration from the official Kyverno GitHub repository (github.com/kyverno) and references standard Sigstore infrastructure (rekor.sigstore.dev) for image verification. These are standard practices for Kubernetes policy management.
[COMMAND_EXECUTION]: The installation section contains common shell commands for Helm and Homebrew to install official Kyverno tools. These are benign and intended for the skill's primary purpose of setting up a policy engine.
[PROMPT_INJECTION]: No evidence of prompt injection, role-play instructions, or safety filter bypasses was detected. The instructions are focused on providing valid YAML policy examples.
[DATA_EXFILTRATION]: There are no patterns suggesting the exfiltration of sensitive data. Hardcoded references to 'registry-credentials' are used as placeholders for Kubernetes secrets, which is a standard configuration pattern.
[INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing Kubernetes resource data. While this represents a surface for indirect injection (processing external YAML), the skill specifically provides policies designed to mitigate such risks by enforcing security standards like 'runAsNonRoot' and 'allowPrivilegeEscalation: false'.

kyverno