ue5-auto-assistant
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, prompt injections, or unauthorized network operations were detected in the skill files.
- [COMMAND_EXECUTION]: The skill architecture explicitly manages access to high-risk capabilities. The mcp-skill-mapping.md file defines guidelines for sub-skills, advising that execute_script and run_console_command should be avoided or used only as a last resort when specialized tools cannot fulfill the request.
- [PROMPT_INJECTION]: The skill acts as a natural language interface for engine operations, which presents a surface for indirect injection.
- Ingestion points: User-provided natural language queries processed according to natural-language-triggers.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for user input interpolation.
- Capability inventory: The system has access to tools for world mutation (spawn_actor), blueprint modification (blueprint_modify), and system-level commands (execute_script).
- Sanitization: The skill mitigates risk through intent-based routing which restricts the available toolset to only what is necessary for the detected task, implementing a form of least privilege.
Audit Metadata