The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of instructions attempting to bypass safety filters or override agent behavior.
[DATA_EXFILTRATION]: No network operations or sensitive data access patterns detected.
[CREDENTIALS_UNSAFE]: No hardcoded secrets or sensitive credentials found.
[OBFUSCATION]: Content is in plain text with no use of Base64 encoding, zero-width characters, or homoglyph attacks.
[EXTERNAL_DOWNLOADS]: Mentions official Godot documentation and a known 2D tool (Laigter), which are reputable sources for the skill's context.
[REMOTE_CODE_EXECUTION]: No patterns of downloading and executing remote scripts or unverified packages were found.
[COMMAND_EXECUTION]: No dangerous shell commands or unauthorized process spawning identified.
[PRIVILEGE_ESCALATION]: No attempts to acquire elevated permissions (e.g., sudo, chmod) were found.
[PERSISTENCE]: No mechanisms for maintaining access across sessions (e.g., modifying shell profiles or cron jobs) were detected.
[DYNAMIC_EXECUTION]: Code examples involve standard Godot functionality such as editor tools (@tool) and shaders, which are expected in this context.
[DYNAMIC_CONTEXT_INJECTION]: No use of shell-command execution at skill load time was found.

godot-prompter-2d-essentials