Back to skill

Security audit

SketchUp-PPT生成

Security checks for vulnerabilities and agentic risk

Overview

This is a local PPT template generator with some overstated SketchUp automation claims, but no evidence of hidden access, persistence, credential use, network activity, or destructive behavior.

Install only if you want a Chinese-language, studio-branded PPT framework generator. It will not truly analyze SketchUp models or capture screenshots, and it may overwrite the output PPTX path you provide. The script also appears to return an undefined variable after saving, so it may report an error even after creating the file.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (8)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The header says the script will read an SU model, extract spatial information, and automatically take screenshots before generating a PPT. In reality, the only model processing is file existence and size/name extraction, and no screenshot or spatial-analysis logic exists anywhere in the file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script's docstrings, usage banner, and runtime messages are written in Chinese, making the skill effectively force a specific language for interaction. Under the policy, locale-specific behavior should either be optional for the user or explicitly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill name and description specify a Chinese-language workflow and organizational template, but do not indicate that users can choose another language or locale. This can violate language/locale policy when a skill implicitly constrains output to a specific language without explicit opt-in or justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

SQP-3 applies to natural-language policy issues in all file types. The document consistently forces a single language locale without indicating user opt-in or a justified locale-specific constraint, which can violate organizational language-choice expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

This is a markdown file, so SQP-2 applies to omissions in user-facing warnings. The usage examples show a command that produces an output .pptx file at a user-specified path, but the documentation does not disclose the file-write behavior or any overwrite risk.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The inline comment suggests these slides correspond to SketchUp screenshots, which implies image extraction or placement. The loop only creates blank slides with titles for default spaces and does not capture, load, or embed any images.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

This Python file performs a filesystem write by saving the generated presentation to the user-supplied output path. Although it logs success after the save, there is no warning beforehand in comments, docstrings, or user-facing output that the command will create or overwrite a PPTX file at that location.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The final message states that the SU file was read successfully, implying meaningful model-level processing. However, the script only checks that the file exists and reads its size and filename, so the message overstates what the code actually did.

Content

No source excerpt is available for this finding.