The Agent Skills Directory

[SAFE]: The skill's instructional content and provided scripts do not contain malicious patterns such as prompt injection, data exfiltration, or persistence mechanisms.\n- [EXTERNAL_DOWNLOADS]: The skill requires standard, well-known Python packages (requests, Pillow) and communicates with official vendor domains (meowa.ai, meowjito.com). These dependencies and network targets are appropriate for the skill's stated purpose and originate from a trusted author context.\n- [COMMAND_EXECUTION]: The provided local tools in the scripts/ directory (e.g., map-preview-server.py, split-components.py) perform legitimate image processing and serve local previews via a loopback HTTP server (127.0.0.1). The server script implements robust security measures, including a restrictive Content Security Policy (CSP) and allowlisting for served media.\n- [SAFE]: API key management instructions adhere to security best practices by guiding users to store credentials in environment variables or .env files, avoiding hardcoded secrets or unsafe prompt-based disclosure.

game-assets