threejs-app-director
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze external application code and assets, which represents a potential surface for indirect prompt injection. This is a standard risk for development-focused skills and is managed through normal agent guardrails.\n
- Ingestion points: Workflow instructions in SKILL.md to inspect existing scene lifecycles, assets, and data.\n
- Boundary markers: None identified in the provided instructions.\n
- Capability inventory: The agent is directed to modify the codebase and perform production builds in SKILL.md.\n
- Sanitization: No specific sanitization of input project files is mentioned.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes external sources, specifically 'Mint MCP' and a 'remote world runtime', to fetch production assets and world manifests for use in the Three.js applications.
Audit Metadata