The Agent Skills Directory

[SAFE]: The skill serves as a reference and SDK guide for developers. It contains no executable scripts or malicious instructions directed at the agent or the user.
[SAFE]: The documentation includes proactive security and privacy guidance for developers, such as emphasizing least privilege, data collection minimization, and mitigating prompt injection risks.
[SAFE]: Code samples provided (e.g., in minimal-mcp-server.md and checkout-flow.md) use safe coding practices, such as using textContent instead of innerHTML to prevent XSS, and using placeholders for sensitive identifiers.
[SAFE]: External tools and services mentioned, such as ngrok and the Model Context Protocol SDKs, are well-known developer utilities. The instructions include appropriate security warnings when using these tools in local environments.

openai-apps-sdk