The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run standard diagnostic shell commands, including ls and npm ls, to detect the project environment. These commands allow the agent to adapt its code generation to match existing project conventions like Tailwind CSS, Jest, or Storybook.
[INDIRECT_PROMPT_INJECTION]: The skill relies on user-supplied component names and project structure data to generate code, which represents an entry point for indirect instructions. (1) Ingestion points: User-provided component names and locations in SKILL.md, and diagnostic output from environment discovery commands. (2) Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat external data as untrusted or to ignore embedded instructions. (3) Capability inventory: The skill has the capability to discover the environment via shell commands and generate new source files. (4) Sanitization: The skill lacks explicit sanitization or validation for user-provided strings before they are used in generated file paths and code templates.

scaffolding-react-components