The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides commands for the user to install additional 'Story Skills' and related tools from GitHub repositories. These are documented for the user to execute manually via the npx skills CLI. The sources include the vendor's own repository (DRincs-Productions/LM-Studio) and other community repositories (danjdewhurst/story-skills, simota/image-skills, sonicfieldlabs/akouo, Leonxlnx/earshot). While these download external content, they are presented as standard installation steps for an agent-based workflow and do not involve silent or malicious execution.
[PROMPT_INJECTION]: The skill contains instructions to follow specific project files and contracts while ignoring examples within the skill for canonical information. These are benign instructional boundaries for an LLM to maintain story consistency and do not constitute malicious injection or safety filter bypasses.
[COMMAND_EXECUTION]: The skill documents shell commands (e.g., npx skills@latest add ...) for the user to run to manage the skill's environment. These are standard CLI usage instructions and do not demonstrate dynamic injection or hidden execution patterns.

narration