Back to skill

Security audit

Motor Drawing Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Chinese-language motor drawing review aid with checklists and a local checklist generator, not a hidden or unsafe agent capability.

Before installing, treat this as a Chinese-language engineering checklist/reference helper rather than an automated CAD or PDF drawing validator. Use generated reports as review aids only, keep human engineering signoff, and use --output only where you intentionally want a text file created.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill description claims it can review actual motor engineering drawings and generate review reports, but the content only shows static checklist/script invocations and reference material, not drawing parsing or automated validation. This mismatch can mislead users into trusting non-existent analysis, causing unsafe engineering decisions or missed defects because a checklist is mistaken for a real review engine.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding

The skill advertises shell commands that write output files (--output review_full.txt) but does not declare any tool scope or permissions boundary. In an agent environment, undeclared file-write behavior can lead to unexpected filesystem modifications, make reviews non-reproducible, and weaken operator control over where data is written.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description and main skill content are entirely in Chinese, and the file does not indicate that language selection is optional or that the skill is intentionally restricted to Chinese-speaking users. Under the policy, imposing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest describes a motor drawing review assistant that audits engineering drawings for pole-slot matching, dimensions, technical requirements, and tolerances, then outputs a review report. In this file, the code merely selects predefined checklist templates and prints or writes them, with no drawing ingestion, validation, or analysis logic.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file contains user-facing instructional content solely in Chinese, and there is no indication that the skill is region-specific or that users can opt into this language. Under the policy rule, forcing a specific language without user choice can constitute a natural-language policy violation.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

The top-level documentation says the tool supports item-by-item confirmation and PDF output, and even shows a --interactive usage example. However, the argument parser only accepts --type and --output, and the implementation only prints plain text or writes a text file.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.