unity-package-management

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill generates local C# scripts (PackageInstaller.cs, PackageSearch.cs, and ProjectSaver.cs) and executes them headlessly using the Unity Editor's -executeMethod flag. This is the standard, documented method for performing asynchronous package operations in a batch-mode or CI/CD Unity environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, specifically the official Unity package registry (packages.unity.com), and local CLI output from the unity tool. While this presents an attack surface where malicious metadata could influence the agent, the risk is mitigated by the use of official, trusted vendor infrastructure.
  • Ingestion points: Metadata fetched from https://packages.unity.com/, output from unity command, and the content of Packages/manifest.json.
  • Boundary markers: The skill instructs the agent to verify IDs against the registry and present a deduplicated list of package IDs to the user for confirmation before installation.
  • Capability inventory: The skill utilizes Bash for CLI operations, file system writing for C# script creation, and subprocess execution to run the Unity Editor binary.
  • Sanitization: The process relies on the built-in validation of the Unity Package Manager (UPM) to resolve dependencies and compatible versions, rather than manual manifest edits.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 03:21 AM