Back to skill

Security audit

code-simplification

Security checks for vulnerabilities and agentic risk

Overview

This looks like a code-simplification skill, but it asks for broad command authority and describes generic automation/API/file workflows that are not clearly scoped to that purpose.

Review before installing. Only use this skill if you are comfortable granting shell command capability to a broadly documented automation/refactoring skill, and restrict it to explicit code-simplification tasks with user approval before any command, credential use, API call, or file mutation.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (5)

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest presents this as a narrow code-clarity refactoring skill, but the body expands scope into generic automation, data processing, workflow orchestration, API use, file handling, and other operational behaviors. That mismatch is dangerous because users or an agent may grant the skill privileges and invoke it under false assumptions, enabling broader actions than its advertised purpose suggests.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

Declaring exec for a skill whose stated purpose is code simplification violates least privilege. In an agent setting, shell execution can be used to run arbitrary commands, access local files, or chain into networked tools, turning a benign refactoring task into a command-execution surface.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

A trigger like 'when the user needs related operations' is so broad that the skill can activate in many unrelated contexts. Combined with the declared exec capability and overbroad scope, ambiguous triggering increases the risk of accidental or opportunistic invocation for tasks the user did not intend this skill to perform.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
80% confidence
Finding

The natural-language description mixes Chinese and English and presents the skill metadata and usage guidance in a fixed bilingual style without indicating that users may choose their preferred language. This can violate language/locale policy when a skill imposes a language presentation format without opt-in or documented justification.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill claims it avoids executing malicious code, yet elsewhere provides command-execution and command-failure workflows, creating an internal contradiction about whether execution occurs. This can mislead reviewers and downstream agents into trusting unsafe behavior, weakening security controls around a capability that can materially affect the host environment.

Content

No source excerpt is available for this finding.