The Agent Skills Directory

[SAFE]: The skill defines standard patterns for improving React component maintainability (hook extraction, component splitting, early returns) without introducing external dependencies or remote execution risks.
[COMMAND_EXECUTION]: The skill utilizes project-local CLI tools such as pnpm analyze-component and pnpm refactor-component, as well as standard development commands like pnpm test, pnpm lint:fix, and pnpm type-check:tsgo. These operations are conducted on local source files and are appropriate for the described development workflow.
[PROMPT_INJECTION]: The skill involves processing local React component source code which constitutes an indirect prompt injection surface. Ingestion points: Source files at user-specified paths are analyzed by pnpm-based tools. Boundary markers: No explicit markers are used to delimit untrusted code from agent instructions. Capability inventory: The skill uses subprocess execution for linting and testing (SKILL.md). Sanitization: No sanitization of the input source code is performed before analysis. The risk is assessed as minimal as the skill focuses on structural metrics and established coding patterns.

component-refactoring