The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill provides capabilities to dynamically generate and write C# source code (.cs files) to the project via dotween_generate_tween_script, dotween_generate_sequence_script, and dotween_generate_lifetime_script. These actions trigger Unity's script compilation and domain reload processes. The skill's own metadata identifies these as high-risk operations due to the execution of the compiler on generated content.
[COMMAND_EXECUTION]: The skill allows for the modification of persistent project assets (such as DOTweenSettings.asset via dotween_settings_configure) and the manipulation of Unity scene components (adding or removing DOTweenAnimation components). These actions directly modify the project environment and editor state.
[INDIRECT_PROMPT_INJECTION]: The skill features a vulnerability surface where instructions embedded in external data could influence generated code.
Ingestion points: The stepsJson parameter in dotween_generate_sequence_script and various configuration parameters across the dotween_pro_* and generation suites.
Boundary markers: The skill description does not mention the use of delimiters or safety instructions to separate data from the code generation templates.
Capability inventory: The skill possesses the ability to write executable files to the file system and modify scene hierarchies.
Sanitization: While basic enum validation and numeric range checks are performed (e.g., validating durations and ease types), the documentation does not indicate sanitization of input strings used in code generation.

unity-dotween