ludeo-unreal-integration

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill automates the acquisition of the LudeoUESDK plugin and the Ludeo C SDK by downloading ZIP archives from 'github.com/ludeo-labs/unreal-plugin-releases' and 'github.com/ludeo-labs/sdk-releases'. These sources belong to the author and represent a well-known service, and are therefore considered safe for development purposes.
  • [COMMAND_EXECUTION]: To automate the Unreal Engine integration, the skill executes several local scripts and batch files (e.g., 'BuildAndPackage.bat', 'RunBPInspector.bat', 'SetupLudeoEnv.ps1'). These scripts invoke standard engine tools like UnrealBuildTool and the Unreal Editor commandlet for compilation, packaging, and Blueprint analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a significant ingestion surface as it reads and analyzes the target game's source code (C++ headers, .uproject files, .Build.cs) and its own internal state file ('.ludeo/integration.json'). While this data is untrusted, the skill uses structured JSON parsing and targeted grep patterns to interpret the codebase, and it implements human-in-the-loop gates for all significant implementation plans.
  • [DYNAMIC_EXECUTION]: The skill utilizes a Python script ('bp_inspector.py') executed via the Unreal Editor's headless commandlet mode to perform automated variable and class introspection. This script is distributed as part of the skill and operates within the context of the editor for introspection purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 07:54 PM
Security Audit — agent-trust-hub — ludeo-unreal-integration