threejs-aaa-graphics-builder
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and analyze external data, including active-play screenshots and performance metrics in JSON format, which creates a surface for indirect prompt injection attacks.
- Ingestion points: The agent is instructed to capture and inspect 'active desktop/mobile screenshots' and read 'metrics and renderBudget blocks' from a JSON inspector output (referenced in SKILL.md and references/visual-scorecard.md).
- Boundary markers: Absent. The skill does not define delimiters or provide instructions to the agent to ignore potentially malicious embedded content within the ingested screenshots or JSON files.
- Capability inventory: The agent has the capability to write and modify project source code (e.g., in src/assets/ and src/systems/), configure rendering pipelines, and invoke other generative AI skills (threejs-3d-generator and threejs-image-generator).
- Sanitization: Absent. There are no mentions of validating, filtering, or sanitizing the external metrics or image data before they are used to inform the agent's graphics-building decisions.
Audit Metadata