The Agent Skills Directory

COMMAND_EXECUTION (SAFE): The skill includes TypeScript build scripts in the build/ directory that perform local file system operations. These operations (reading markdown files and writing the compiled AGENTS.md document) are restricted to the project directory and are standard for documentation build tools.
EXTERNAL_DOWNLOADS (SAFE): The project utilizes standard development dependencies such as typescript and tsx as listed in build/package.json. No suspicious external downloads, piped remote script executions (e.g., curl | bash), or unverified packages were identified.
PROMPT_INJECTION (SAFE): The instructions for AI agents are strictly focused on WordPress performance optimization. There are no attempts to bypass safety filters, extract system prompts, or override agent behavioral constraints.
DATA_EXFILTRATION (SAFE): No code patterns or commands were found that access sensitive system files (e.g., .ssh, .env) or attempt to send data to external domains. The provided PHP examples adhere to security best practices, specifically promoting the use of prepared statements and nonces to prevent SQL injection and CSRF.

wordpress-performance-best-practices