playcanvas-engine

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides Python automation utilities located in the scripts/ directory (component_builder.py and project_generator.py). These tools are designed to scaffold JavaScript components and HTML project files by creating directories and writing generated code to the local file system using the Python standard library.
  • [EXTERNAL_DOWNLOADS]: The starter template and various examples in the documentation fetch the PlayCanvas engine and the Ammo.js physics library from established and well-known content delivery networks, specifically code.playcanvas.com and cdn.jsdelivr.net.
  • [INDIRECT_PROMPT_INJECTION]: The project contains scripts that accept user-provided strings for file names and directory paths to automate project setup. While this represents a potential surface for path traversal if an agent were manipulated into using malicious input from an untrusted external source (such as a pull request description), the capability is limited to local file generation consistent with the skill's primary purpose as a developer tool.
  • Ingestion points: Command-line arguments and standard input calls in scripts/component_builder.py and scripts/project_generator.py.
  • Boundary markers: None identified in the script logic or documentation.
  • Capability inventory: File system writes and directory creation across all automation scripts.
  • Sanitization: Basic alphanumeric validation is performed on component names in interactive mode, although path sanitization for output directories is minimal, which is typical for developer-facing CLI tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 07:37 AM
Security Audit — agent-trust-hub — playcanvas-engine