The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains deceptive metadata regarding its authorship. In SKILL.md, it identifies 'Anthony Fu' as the author, whereas the skill was provided by 'marcioaltoe'. This misattribution is misleading concerning the provenance and official nature of the skill content.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to process untrusted configuration data.
Ingestion points: The agent is guided to read and implement settings from vite.config.ts, .env, and project source files.
Boundary markers: There are no instructions or patterns provided to isolate untrusted configuration data or to ignore instructions potentially embedded in these files.
Capability inventory: The skill enables the agent to execute shell commands (vite, vite build) and perform file system operations based on the processed configuration.
Sanitization: The instructions do not include validation or sanitization steps for configuration values before they are used to drive agent actions.
[COMMAND_EXECUTION]: The skill documents and facilitates the execution of CLI commands such as vite, vite build, and vite preview for development and production build tasks.
[EXTERNAL_DOWNLOADS]: The skill references the installation of standard Node.js packages from the NPM registry, including vite, rolldown-vite, and various official Vite plugins. These references target well-known repositories.

vite