The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains instructional constraints using 'MUST' and 'CRITICAL' tags. These are used to enforce specific software design patterns and coding conventions (e.g., component state management, prop forwarding) and do not attempt to bypass safety filters or override the agent's core instructions.
[DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive file paths, environment variable access, or hardcoded credentials were found. The skill does not perform any network operations to external or untrusted domains.
[OBFUSCATION]: The content is clear and uses standard Markdown and TSX. No hidden characters, encoded strings (Base64/Hex), or homoglyphs were identified.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts. References to external libraries (e.g., Radix UI, Base UI) are limited to standard package imports for component development purposes.
[INDIRECT_PROMPT_INJECTION]: As a code refactoring and design skill, it possesses an ingestion surface (processing user-provided component code). However, it includes strict boundary markers and architectural rules that guide the agent toward safe, predictable coding practices. The risk is assessed as safe/low.
[DYNAMIC_EXECUTION]: The skill uses standard React patterns like React.cloneElement and context providers. It does not employ unsafe dynamic execution methods like eval(), exec(), or runtime compilation of untrusted input.

meta-design-composable-components