The Agent Skills Directory

[SAFE]: The skill serves as a configuration assistant for DevOps tasks related to mirrord and Temporal. It includes comprehensive security boundaries that prevent the AI agent from handling sensitive credentials directly, instead directing users to use Kubernetes Secrets. The skill also enforces input validation and restricts cluster operations to read-only discovery, ensuring that any changes to the environment must be reviewed and executed by the user manually.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user inputs and Kubernetes resources (e.g., container names, env vars, and deployment YAMLs) to generate configuration and run discovery commands.
Ingestion points: Data is pulled from the cluster via kubectl get deployment/<name> -o yaml (SKILL.md, Step 2) and from user-provided names and filters.
Boundary markers: The agent is instructed to treat user input as data and never as instructions, and to use delimiters for external content.
Capability inventory: The skill uses kubectl to read cluster state and generates YAML files. It does not perform automated write operations.
Sanitization: The skill explicitly requires validating Kubernetes names against regex (^[a-z0-9]([a-z0-9-]{0,61}[a-z0-9])?$) and rejecting shell metacharacters before use in commands.

mirrord-temporal