The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use several local CLI tools to manage the desktop environment, including quickshell for UI rendering, hyprctl for window manager configuration, and rg (ripgrep) for searching configuration files within the user's home directory. It also uses git for staging specific hunks of code changes.
[DYNAMIC_EXECUTION]: Employs a specific Python one-liner (python3 -c) to verify font glyph names directly from the font file (FiraCodeNerdFont-Regular.ttf). This is a defensive check to ensure icons render correctly. It also defines a dumpUi JavaScript function within the QML context to inspect the UI tree for layout verification.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user configuration files and potentially IPC (Inter-Process Communication) targets. However, it mitigates risks by mandating 'Defensive QML' practices, such as wrapping JSON.parse in try/catch blocks and validating the shape of data before use.
[SAFE]: The workflow uses xdg-open to launch local HTML previews in the user's default browser, which is a standard and safe method for visual verification. No external network requests or suspicious remote code patterns were identified.

quickshell-modules