unity-cinemachine

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill documentation indicates the use of a runtime reflection adapter ('CinemachineAdapter' / 'CinemachineSkills') to bridge API differences between Cinemachine versions 2.x and 3.x. This allows the tool to dynamically resolve and set properties like 'propertyName' at runtime based on the detected environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted string data for identifying and configuring GameObjects and properties, which creates a potential surface for indirect prompt injection if the agent is processing data from untrusted sources.
  • Ingestion points: Parameters in 'SKILL.md' such as 'vcamName', 'followName', 'lookAtName', and 'propertyName'.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content within scene object names are defined in the provided metadata.
  • Capability inventory: The skill allows for the creation, modification, and deletion of Unity components and GameObjects, as well as setting arbitrary properties via reflection.
  • Sanitization: The skill performs basic semantic validation for specific enumerated fields like 'mode', but general property access is dictated by string identifiers provided at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 06:52 AM