The Agent Skills Directory

[PROMPT_INJECTION]: The instructions are strictly focused on UI quality, accessibility, and adherence to design systems. No attempts to bypass safety filters or override agent behavior were detected.
[DATA_EXFILTRATION]: No hardcoded credentials, sensitive file access, or unauthorized network operations were found. The skill does not communicate with external domains.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. It relies on local scripts for project-specific auditing.
[COMMAND_EXECUTION]: The skill includes Node.js and Python scripts (check-contrast.mjs, detect-token-drift.mjs, validate-metadata.py) to perform design-related checks. These are used locally and do not involve untrusted inputs.
[OBFUSCATION]: No evidence of Base64 encoding, hidden characters, or other obfuscation techniques was found in the documentation or scripts.
[DYNAMIC_EXECUTION]: No evidence of runtime code generation or unsafe deserialization of data was detected.
[DATA_EXPOSURE]: While the skill reads local project files (e.g., DESIGN.md, tokens.css), this is intended for its primary purpose of UI auditing and does not target sensitive system configuration files.

ui-craft