polyhaven-texture-apply
Fail
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/polyhaven-texture-apply.jscontains a command injection vulnerability due to unsafe string interpolation in theexecute_codefallback logic. - The variables
objectName,matName, andscale, which are derived from command-line arguments provided to the skill, are directly embedded into a Python script template. - An attacker could provide a crafted object name such as
MyMesh"); import os; os.system("malicious_command"); #to escape thebpy.data.objects.get()function call and execute arbitrary Python and system commands. - [REMOTE_CODE_EXECUTION]: By generating and sending unvalidated Python code to the Blender MCP server, the skill creates a mechanism for arbitrary code execution on the user's host machine.
- [DATA_EXFILTRATION]: The skill lacks sanitization and boundary markers for its ingestion points (
--object,--texture,--scale). This exposure, combined with the capability to execute code, could be leveraged to exfiltrate local files or sensitive data from the Blender environment.
Recommendations
- AI detected serious security threats
Audit Metadata