Skills
skills/besty0728/unity-skills/unity-shader/Gen Agent Trust Hub

unity-shader

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface.
  • Ingestion points: The shader_read function ingests content from files specified by the shaderPath parameter.
  • Boundary markers: None; there are no specified delimiters or instructions to ignore instructions embedded within the shader source code.
  • Capability inventory: The skill includes shader_create (file write), shader_delete (file deletion), and shader_set_global_keyword (global state modification).
  • Sanitization: None; the documentation does not describe path validation logic to ensure operations are restricted to the Unity project directory.
  • [DATA_EXFILTRATION]: The shader_read tool allows reading of file content. If the implementation lacks strict path validation (e.g., scoping to the Assets folder), it could be exploited to read sensitive local files if the agent is misled via prompt manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 11:28 PM