babylonjs-engine
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill includes Python scripts (
scripts/mesh_builder.pyandscripts/scene_generator.py) that generate JavaScript and TypeScript code based on user-supplied parameters. These scripts use predefined templates to assemble code snippets, which is a standard utility for developer tools. - [INDIRECT_PROMPT_INJECTION]: The code generation scripts represent a potential surface for indirect prompt injection as they ingest untrusted user input to generate executable code.
- Ingestion points: User-provided CLI arguments and JSON parameters in
scripts/mesh_builder.pyandscripts/scene_generator.py. - Boundary markers: The scripts do not use explicit delimiters or "ignore embedded instructions" markers in the generated output files.
- Capability inventory: The scripts have the capability to write generated files to the local filesystem using
pathlib.Path.write_text(). - Sanitization: The scripts perform basic type validation for numerical and boolean inputs but do not fully sanitize string inputs before interpolating them into code templates.
- [EXTERNAL_DOWNLOADS]: Multiple files (e.g.,
SKILL.md,assets/examples/README.md) contain instructions to fetch 3D models and environment textures fromhttps://assets.babylonjs.com/. This is the official, well-known asset repository for the Babylon.js engine and is considered a safe and standard source. - [COMMAND_EXECUTION]: The
assets/starter_babylon/package.jsonfile defines standard Node.js scripts (dev,build,preview) that execute thevitebuild tool. This is expected behavior for a web development starter template.
Audit Metadata