The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill utilizes shell commands (node -v, npm -v) and inline Node.js scripts (node -e) exclusively for environment and framework detection. These operations are standard for development tools and do not involve untrusted inputs or high-privilege operations.
[DATA_EXFILTRATION] (SAFE): Filesystem access is limited to standard project configuration files such as package.json, pnpm-lock.yaml, and yarn.lock. No sensitive user data, credentials, or system files are accessed or transmitted.
[REMOTE_CODE_EXECUTION] (SAFE): No remote script downloads (e.g., via curl or wget) or execution of untrusted remote code were detected. The skill does not attempt to install any external packages.
[PROMPT_INJECTION] (SAFE): No instructions designed to override agent safety protocols or ignore previous guidelines were found in the markdown or metadata.

nodejs-expert