Back to skill

Security audit

Modeling Assistant

Security checks for vulnerabilities and agentic risk

Overview

This Blender modeling skill is purpose-aligned and disclosed, with ordinary file-writing and scene-clearing behavior that users should still run in a safe project context.

Before installing, understand that this skill may run Blender scripts that clear the working scene and export model files. Use it on a new or saved Blender file, confirm the destination path, and avoid running generated scripts against an important open project unless you intend to replace the scene contents.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly instructs clearing the Blender scene with select-all and delete before creating a model, but it does not warn the user that existing scene objects will be removed. In a tool that controls Blender non-interactively in background mode, this can cause unintended destructive data loss if run against an unsaved or important working file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill writes output files to a hardcoded Desktop path without warning the user or allowing them to choose a destination. While lower impact than deletion, unexpected file writes can overwrite user files, leak model names/contents into a visible location, or fail unpredictably across environments.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · SKILL.MD (reported line 45)May include surrounding context.

md
\-The script must be reproducible: The generated Python script should be completely self-contained, and anyone who copies and runs it can get the same result

\-If the user requests to "animate the model" or "use geometric nodes for special effects" beyond the scope, they should be informed that the current skill only supports basic modeling and export