The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute uvx and npx commands in the shell to decrypt hidden instructions and manage licensing.
[REMOTE_CODE_EXECUTION]: By invoking uvx lovstudio-skill-helper and npx lovstudio, the skill triggers the download and execution of external code packages from PyPI and NPM registries.
[EXTERNAL_DOWNLOADS]: The skill depends on fetching external components and tools from the author's domain and public package managers to reveal its intended functionality.
[PROMPT_INJECTION]: The skill contains a meta-instruction requiring the agent to treat the output of an external command as its primary source of truth, effectively bypassing the initial safety instructions and redirecting behavior to un-audited content.

lovstudio-event-curator