blender-mcp
Fail
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions include a command to download a Python script (
addon.py) from an external repository (raw.githubusercontent.com/ahujasid/blender-mcp/main/addon.py). The source repository belongs to an untrusted account. - [REMOTE_CODE_EXECUTION]: The skill facilitates the installation of a socket server that allows an external agent to send and execute code on the host machine. This architecture creates a remote code execution capability by design.
- [DYNAMIC_EXECUTION]: The primary function of the skill is the
execute_codecommand, which allows for the execution of arbitrary Python code within the Blender process. This environment has full access to standard Python libraries, enabling unrestricted command execution on the host machine. - [COMMAND_EXECUTION]: The skill instructs the user to execute shell commands like
curlto download external content andnc(netcat) to check local port status, which involves direct interaction with the host operating system.
Recommendations
- AI detected serious security threats
Audit Metadata