The Agent Skills Directory

[COMMAND_EXECUTION]: The skill features multiple Bash, Python, and Groovy script templates that utilize CLI tools like kubectl, terraform, docker, and the GitHub CLI (gh). These scripts are used for deployment automation, infrastructure provisioning, and forensic evidence collection. While powerful, they are contextually appropriate for the defined senior DevOps role and contain best-practice constraints.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external software supply chains, creating an inherent attack surface for indirect prompt injection.
Ingestion points: The skill interacts with external Git repositories, CI/CD event payloads, container registry metadata, and application logs processed during automated incident response.
Boundary markers: Explicit markers or instructions to isolate untrusted external data from agent commands are not defined within the provided script templates.
Capability inventory: The skill has significant capabilities, including full infrastructure modification via Terraform, container management via Kubernetes and Docker, and network communications via curl and specialized APIs.
Sanitization: The code snippets demonstrate standard automation logic but do not include explicit input validation or sanitization routines for data ingested from external sources.

devops-engineer