environment-shot
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process an external 3D brief provided by an upstream skill.
- Ingestion points: The skill reads a 'brief' from
intent-to-3d-brief-writeras defined inSKILL.mdandcontract.json. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt interpolation logic.
- Capability inventory: The skill has zero execution capabilities; it performs no subprocess calls, file writes, or network operations.
- Sanitization: No validation or sanitization of the input brief content is performed beyond checking if the locale scope is 'realizable'.
- [SAFE]: The skill incorporates strong security boundaries via its
contract.jsonwhich enforces atruthStatePolicyofNot Runand amutationRuleofdowngrade_only, ensuring the agent cannot elevate the skill's capabilities to execute code or Blender commands.
Audit Metadata