The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to perform research by scanning the Omarchy marketplace catalog and cloning third-party plugins, which introduces untrusted content into the execution context.\n
Ingestion points: SKILL.md (Phase 1: Market scan via catalog.json and shallow-cloning of 3 to 5 nearest plugins).\n
Boundary markers: Absent. The instructions do not include specific delimiters or warnings to ignore instructions that might be embedded in these external repositories.\n
Capability inventory: The skill utilizes significant capabilities across its phases, including repository creation (gh repo create), plugin installation/execution (omarchy plugin add), and system UI automation (hyprctl, grim).\n
Sanitization: Absent. No sanitization or validation of the content retrieved from the marketplace or cloned repositories is mentioned before processing.\n- [COMMAND_EXECUTION]: The skill makes extensive use of CLI tools for automation, testing, and UI verification.\n
Evidence: Uses hyprctl eval to programmatically move the system cursor and grim to capture screenshots of the desktop for verification purposes.\n
Evidence: Employs the omarchy CLI for plugin lifecycle management and gh for repository operations and marketplace submission.\n
Evidence: Instructions include using notify-send for desktop notifications and a Python-based HTTP server for local prototype preview.\n- [EXTERNAL_DOWNLOADS]: The skill references several external GitHub repositories for templates and documentation.\n
Evidence: Recommends cloning github.com/steveclarke/omarchy-headroom as a project template. (Note: This is a vendor resource owned by the skill author).\n
Evidence: References documentation and tools from github.com/tcballard/build-omarchy-plugins and github.com/omacom/omarchy-plugin-marketplace.

omarchy-plugin