The Agent Skills Directory

[SAFE]: No security issues detected across any of the 11 threat categories. The skill is entirely composed of documentation and shader code templates.
[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety filters, or extract system prompts were found in the markdown files or code comments.
[DATA_EXFILTRATION]: There is no evidence of commands targeting sensitive file paths (e.g., .ssh, .env, .aws) or performing unauthorized network operations. All code is restricted to local graphics rendering and math.
[OBFUSCATION]: No Base64-encoded executable strings, zero-width characters, or homoglyph attacks were detected in any of the analyzed files.
[REMOTE_CODE_EXECUTION]: The skill does not attempt to install external dependencies via npm or pip, nor does it perform remote script execution patterns like 'curl | bash'.
[DYNAMIC_CONTEXT_INJECTION]: The SKILL.md file does not contain any dynamic shell execution placeholders using the '!command' syntax.

shader-dev