If you discover a security issue in this repository — for example a skill that could induce unsafe code generation, leak data, or bypass a runtime sandbox — please report it privately.

• Preferred: open a GitHub security advisory (private disclosure).
• Email: security@agenticwork.io

Please include a description, reproduction steps, and the potential impact. Do not open a public issue for a security-sensitive report.

We aim to acknowledge reports within a few business days and will coordinate a fix and disclosure timeline with you.

This repository contains skills (instructional Markdown + optional resources/scripts). The primary risk surface is a skill that steers a model toward unsafe output. Skills here must assume their output runs in a hardened sandbox (opaque-origin iframe, pinned imports only, no ambient credentials) and must never instruct a model to weaken those controls.