The Agent Skills Directory

[SAFE]: The skill is designed for local auditing of GitOps repositories and does not perform any unauthorized data exfiltration or access sensitive system paths outside of the specified repository root.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute bundled scripts (discover.sh, validate.sh, check-deprecated.sh). These scripts are transparent and use standard industry utilities such as awk, git, kustomize, kubeconform, and yq. These operations are consistent with the skill's stated purpose.
[PROMPT_INJECTION]: As the skill ingests and analyzes user-provided manifest files, it is subject to the risk of indirect prompt injection if those files contain text intended to influence the agent's conclusions. This risk is inherent to any analysis task and is categorized as low severity given the use-case. Implementation of boundary markers when presenting file content to the LLM is recommended as a mitigation.

gitops-repo-audit