The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a bash utility, find-polluter.sh, which automates the identification of state-polluting tests by executing files sequentially via npm test. Additionally, SKILL.md provides examples of diagnostic shell commands such as env, security list-keychains, and codesign to help identify environmental discrepancies in build pipelines. These commands are standard for development and troubleshooting workflows.
[PROMPT_INJECTION]: The skill describes techniques for analyzing external data such as error messages, logs, and stack traces, which constitutes an indirect prompt injection surface. Ingestion points: Error messages, stack traces, and logs ingested during Phase 1 (Root Cause Investigation). Boundary markers: None present in the provided instructions. Capability inventory: Local command execution (shell) and test runner execution (npm) as demonstrated in the diagnostic examples and find-polluter.sh. Sanitization: No explicit sanitization or escaping techniques are detailed for the processed log data, as the skill focuses on technical analysis rather than automated data handling.

systematic-debugging