STOP. READ THIS ENTIRE SKILL.MD BEFORE CALLING ANY ENDPOINT.

Security Best Practices

Use this skill to define what "secure enough to proceed" means before running /hack, /security-scan, /code-runner proof probes, or a manual code review. The goal is authorized hardening: find exploitable weaknesses, validate them in bounded local containers, and produce patch/regression-test plans.

Seed Sources

Use these as canonical external baselines when /dogpile needs fresh context: