Back to skill

Security audit

数据分析理调

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly a data-analysis guidance document, but it declares write and command-execution capabilities that are not justified and are partly contradicted by its own text.

Review before installing. The content does not show malicious behavior, but the declared permissions are stronger than the documented purpose. Prefer a version that removes exec and write, or clearly documents the exact commands, file changes, user approvals, and safeguards required for those capabilities.

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 (4)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation explicitly classifies the skill as pure Markdown with no need for exec, while the manifest grants exec capability. This mismatch can mislead users and reviewers into trusting a skill with stronger privileges than its documented behavior, increasing the risk of unintended command execution if the skill is later extended or abused.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill is presented as a methodology and guidance framework, yet it declares exec capability alongside read/write/glob. For a primarily instructional analysis skill, granting command execution creates unnecessary attack surface and could enable filesystem or shell actions unrelated to the stated purpose.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
72% confidence
Finding

The invocation guidance is broad and open-ended, encouraging use across many data-analysis and workflow scenarios without narrowing boundaries or clarifying when powerful tools may be invoked. Ambiguous scope can cause overuse of a high-privilege skill in contexts where users do not expect file modification or command execution.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill declares read, write, exec, and glob capabilities but does not provide a clear upfront warning that it may alter files or execute commands. Users may invoke it expecting analytical guidance only, creating a consent and transparency gap that makes misuse or accidental system impact more likely.

Content

No source excerpt is available for this finding.