The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents and utilizes several command-line tools and local scripts for operational tasks, including kubectl for log retrieval, git for versioning, psql for database connectivity testing, and custom shell scripts such as deploy.sh, run_migrations.sh, and switch_traffic.sh for deployment orchestration.
[EXTERNAL_DOWNLOADS]: The included health check script (scripts/health_check.py) uses the requests library to perform connectivity tests against configured API endpoints using well-known placeholder domains (e.g., api.example.com).
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection during troubleshooting operations. (1) Ingestion points: The agent is instructed to read and diagnose application error logs via kubectl logs and tail in the troubleshooting section of SKILL.md. (2) Boundary markers: No specific delimiters are defined to isolate untrusted log data from agent instructions. (3) Capability inventory: The skill possesses the ability to execute various deployment and traffic-switching shell scripts and perform network operations. (4) Sanitization: There is no explicit sanitization step for the content of logs before agent analysis.

deployment-runbook