The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill leverages the unity command eval tool to execute dynamically generated C# statement blocks within a live Unity Editor. While the skill provides specific templates for graphics configuration in SKILL.md and references/code-templates.md, this mechanism allows for arbitrary C# execution in the target environment.
[INDIRECT_PROMPT_INJECTION]: The skill translates high-level user requests (e.g., cinematic or horror visual styles) into executable code snippets. This design creates a vulnerability surface where adversarial user input could attempt to influence the logic of the generated C# code.
Ingestion points: User queries regarding post-processing effects, bloom, tonemapping, and other volume overrides (referenced in SKILL.md).
Boundary markers: The templates do not employ explicit boundary markers or delimiters to isolate user-provided values from the code logic.
Capability inventory: The skill utilizes the unity command eval capability to perform file system operations via AssetDatabase, object modification via Undo, and runtime state changes via the Volume framework (referenced in references/code-templates.md).
Sanitization: There is no evidence of sanitization or validation of user-provided parameters before they are interpolated into the C# snippets.

urp-postprocessing