The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides methods to read and process data from external Unreal Engine assets (Niagara Systems). This creates a potential attack surface if the assets contain untrusted data designed to influence the agent's behavior.
Ingestion points: Data is ingested via unreal.NiagaraService.list_rapid_iteration_params, unreal.NiagaraService.compare_systems, and unreal.NiagaraService.get_emitter_lifecycle as described in SKILL.md and scripts/create_system.txt.
Boundary markers: No specific delimiters or safety instructions are defined to separate asset data from the agent's command context.
Capability inventory: The skill utilizes execute_python_code and call_tool for engine interaction and file operations.
Sanitization: The instructions do not specify any validation or sanitization routines for the data retrieved from the Niagara assets.
[DYNAMIC_EXECUTION]: The skill's primary functionality relies on the execute_python_code tool to run dynamically generated Python scripts that interface with the unreal module. While this is the intended use case for editor automation, it involves the execution of code generated by the agent at runtime.

niagara-systems