The Agent Skills Directory

[SAFE]: The skill implements authentication and authorization logic using industry-standard libraries and patterns. It explicitly includes security-hardening instructions such as using set search_path = '' for PostgreSQL security definer functions to prevent search path attacks.
[SAFE]: Instructions for data handling prioritize security, directing the agent to use server-side validation, avoid trusting client-side session data as the sole authorization boundary, and verify webhook signatures to prevent forgery.
[SAFE]: The skill promotes the Principle of Least Privilege by advising against redundant service-role policies and ensuring Row Level Security (RLS) is applied to all data access points.
[SAFE]: Environment variable management follows security best practices, instructing the agent to keep secret keys server-only and use placeholder files for public configuration.

auth-implementation