The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes find-polluter.sh, a bash utility that executes npm test on files discovered through a user-provided search pattern to identify which tests cause environmental or file state pollution.
[COMMAND_EXECUTION]: Guidelines in SKILL.md and root-cause-tracing.md instruct the agent to use system CLI tools such as git, codesign, and security to collect environment metadata and troubleshoot build or signing failures.
[PROMPT_INJECTION]: The skill creates a potential surface for indirect prompt injection because its core functionality requires ingesting and analyzing untrusted external data sources such as application logs, error messages, and stack traces. Ingestion points: Error messages and stack traces are analyzed in SKILL.md and root-cause-tracing.md. Boundary markers: No explicit delimiters are specified for the isolation of external log content. Capability inventory: The skill utilizes command execution for testing (npm test) and system diagnostics (git, codesign). Sanitization: The instructions do not define specific sanitization or filtering logic for processed logs.

systematic-debugging