The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes bundled shell scripts (scripts/discover.sh, scripts/validate.sh, scripts/check-deprecated.sh) to automate the discovery and validation of Kubernetes resources. These scripts invoke external binaries such as flux, kustomize, yq, and kubeconform. All operations are performed locally on the repository provided by the user.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it is designed to ingest and analyze untrusted third-party Kubernetes manifests. Adversarial content within a repository could attempt to influence the auditor's findings or recommendations.
Ingestion points: Local repository manifests processed during discovery, validation, and security review phases as described in SKILL.md.
Boundary markers: Absent; the skill does not explicitly use delimiters or instructions to ignore embedded prompts within the files it reads.
Capability inventory: Execution of shell scripts and standard GitOps tooling; filesystem read access to the target repository; audit report generation.
Sanitization: Manifests are validated against strict OpenAPI schemas using kubeconform to ensure structural integrity prior to analysis.

gitops-repo-audit