The Agent Skills Directory

[SAFE]: The skill enforces best practices for financial security and user consent by implementing an explicit cost-transparency gate. It requires the agent to display the estimated credit cost and current balance, pausing for user confirmation before executing any paid tool calls.\n- [SAFE]: Visual assets and dependencies are sourced exclusively from trusted or well-known services. The skill utilizes Google Fonts via official 'fonts.gstatic.com' endpoints for typography and leverages the vendor's own infrastructure for asset hosting, ensuring no exposure to untrusted third-party domains.\n- [SAFE]: The rendering pipeline uses a server-side HTML-to-PNG process with JavaScript disabled by default. This architecture minimizes the risk of local command execution or client-side script injection while maintaining high-fidelity output.\n- [SAFE]: Quality assurance is integrated into the workflow, using media analysis to inspect rendered assets for compliance with platform safe-zones and brand guidelines before final delivery to the user.\n- [SAFE]: The skill documentation provides clear push-back criteria to prevent the generation of generic or hallucinated content, ensuring the tool remains grounded in the user's real product data and brand specifications.

app-store-screens