The Agent Skills Directory

[SAFE]: The skill implements secure secret management by instructing the agent to never request secrets via chat and instead facilitates local entry into .env files with restricted filesystem permissions (chmod 0600).
[SAFE]: The architecture utilizes well-known and reputable services and libraries, such as Supabase for Postgres, official Docker images for accessories (Redis, n8n, Meilisearch), and standard Go/Node.js tooling.
[SAFE]: It includes a development-only authentication bypass endpoint guarded by environment variables (DEV_MODE), with explicit instructions to ensure this functionality is physically excluded from production binaries during route registration.
[SAFE]: The skill promotes security best practices, such as mandatory server-side validation, authorization checks in every handler, and discouraging insecure authentication methods for production use.

cofounder-tech-stack