The Agent Skills Directory

[CREDENTIALS_UNSAFE] (SAFE): No hardcoded production credentials detected. Code snippets use environment variables (os.getenv, process.env) and provide clear warnings against using default development keys in production.
[EXTERNAL_DOWNLOADS] (SAFE): No suspicious external downloads or remote script execution patterns identified. All mentioned dependencies are standard industry libraries.
[REMOTE_CODE_EXECUTION] (SAFE): No patterns of remote code execution or dynamic code evaluation (eval, exec) found.
[DATA_EXFILTRATION] (SAFE): No unauthorized data access or exfiltration patterns detected. Network operations are limited to standard OAuth 2.0 flows to trusted providers (Google).
[PROMPT_INJECTION] (SAFE): No instructions attempting to override agent behavior or bypass safety filters were found.
[COMMAND_EXECUTION] (SAFE): No arbitrary command execution or subprocess spawning detected.

api-authentication