The Agent Skills Directory

Log Sanitization and Sensitive Data Protection: The skill documentation outlines a robust defense-in-depth approach for protecting sensitive information. It introduces utility functions such as redact_token, sanitize_url, and sanitize_params, alongside a global SensitiveDataFilter. These mechanisms are designed to identify and redact passwords, tokens, API keys, and JWTs from server logs, which is a proactive security practice for preventing data exposure.
Structured Error Reporting: The skill enforces a consistent protocol for API responses, distinguishing between success, validation errors, and authentication failures. By utilizing structured error bodies and a global error handler, it helps ensure that raw backend exceptions and internal stack traces are not leaked to the frontend, thereby reducing the risk of information disclosure.
Auth Error Handling: The protocol explicitly handles authentication and authorization errors (401 and 403 status codes), ensuring that these security-relevant events are correctly categorized and communicated to the frontend client through dedicated handlers.

error-handling