threejs-debug-profiler
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest external asset URLs and remote runtime configuration (specifically the "Mint world RAD URL"). This creates a surface where remote, potentially untrusted data can influence the agent's diagnostic logic and subsequent modifications to the project's source code.
- Ingestion points: Remote glTF/asset URLs and the "Mint world RAD URL" mentioned in references/debug-profile-checklists.md.
- Boundary markers: No instructions are provided to the agent to treat external content as untrusted or to ignore embedded instructions within these data sources.
- Capability inventory: The agent is tasked with modifying project code based on its findings ("Fix the cause in the owning module").
- Sanitization: The skill lacks requirements for validating or sanitizing the content of the remote configuration before it is used to guide code changes.
Audit Metadata