The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides automated scripts for incident response, platform engineering, and release management that interact directly with infrastructure components using kubectl, terraform, docker, and the GitHub gh CLI.
[DATA_EXFILTRATION]: Includes a forensic evidence collection script (collect-evidence.sh) that gathers sensitive diagnostic data, including system logs, database activity (pg_stat_activity), and network packet captures (tcpdump), though these are stored in a local directory for analysis.
[INDIRECT_PROMPT_INJECTION]: The skill processes complex configuration data, which presents an attack surface for indirect prompt injection.
Ingestion points: Deployment manifests, GitHub Action workflows, and Terraform HCL templates in references/ and SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the templates.
Capability inventory: Extensive system access via kubectl, terraform, docker, and the gh CLI across multiple reference scripts.
Sanitization: No explicit sanitization or validation logic is defined for data interpolated into configuration templates.

devops-engineer