The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides installation commands for the Valet CLI tool which involve downloading and executing shell and PowerShell scripts from the vendor's domain (https://valet.dev/install.sh and https://valet.dev/install.ps1). These are official distribution channels for the vendor's software.
[COMMAND_EXECUTION]: The agent is instructed to use the valet command-line utility for operations such as site creation, file deployment, and downloading site assets. This is the primary intended functionality of the skill.
[EXTERNAL_DOWNLOADS]: The skill fetches the Valet CLI installer from the developer's infrastructure and recommends updating the skill via npx from the official valetdotdev GitHub repository.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and publish local project files which could contain untrusted data.
Ingestion points: Project files, directories, and generated artifacts provided for publication (SKILL.md).
Boundary markers: The skill provides explicit instructions to use dedicated scratch directories and to list/verify directory contents before deployment to prevent the accidental upload of sensitive files.
Capability inventory: The skill utilizes shell execution for the valet CLI and performs network operations to upload files to the hosting service.
Sanitization: Files are published as-is to provide static hosting; the skill guides users to manually remove sensitive files like .env and .git prior to deployment.

valet-publish