The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell and PowerShell scripts to collect system performance data using standard utilities like ps, top, lsof, and Get-Process. It also includes gated functionality to stop user-level processes using gentle signals like TERM.
[DATA_EXPOSURE]: It gathers detailed system telemetry but implements strong privacy measures, such as defaulting to process names instead of full command lines to avoid secret exposure and using regex-based redaction in scripts/compare_snapshots.py to filter out credentials from reports.
[SAFE]: The skill operates entirely within the local environment without external network dependencies. It defines strict 'Safety Rules' and a 'Dangerous Action Gate' to prevent unauthorized system changes or privilege escalation.

optimize-system-performance