The Agent Skills Directory

[SAFE]: The skill is designed to provide instructional content regarding the es-toolkit library. It uses local tools such as Read, Grep, and Glob to access documentation files provided within the skill package.
[EXTERNAL_DOWNLOADS]: The skill references installation via standard package managers (npm, yarn, pnpm, bun, and deno) for the es-toolkit library. These are provided as informational snippets for the user and are not executed by the agent or skill environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from local documentation files (e.g., docs/*.md) to answer user queries. While this represents an ingestion surface for indirect instructions, the skill environment is restricted to read-only operations and lacks capabilities for network access, file writing, or command execution, mitigating the risk of exploitation.

guide