Back to skill

Security audit

data-scientist

Security checks for vulnerabilities and agentic risk

Overview

This package looks like a data-science skill but mostly contains governance and licensing text while requesting broad read, write, and web-search tools, so it should be reviewed before installation.

Install only if you are comfortable with a broadly triggered, documentation-heavy data-science skill that may receive file read/write and web-search authority. Prefer a version that either provides real scoped data-analysis instructions or removes broad tools and generic triggers.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill is presented in the manifest as a Python data analysis and visualization capability, but the document later states the package contains no code and is purely documentation/theoretical text. This mismatch can mislead agents or operators into granting capabilities, selecting the skill for execution, or trusting outputs under false assumptions about implementation and review status.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 91)May include surrounding context.

md
| 文档与理论文本 | 本 `SKILL.md`、`README.md`、其中「理论依据」段与一切理论表述 | **不在 MIT 覆盖范围内**:保留所有权利(All rights reserved) |

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest declares powerful tools such as Write and WebSearch even though the package later claims to contain no executable skill code and only documentation. Over-declared permissions enlarge the effective attack surface because an agent platform may authorize file modification or external lookup based solely on metadata, enabling unintended actions by a mislabeled or placeholder skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest lists trigger keywords only as "data" and "scientist", which are common terms and do not clearly distinguish when this skill should activate versus when it should not. The file also lacks scope constraints or negative examples to prevent unintended invocation collisions.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The manifest description embeds a very broad trigger list including generic phrases such as “可视化/画图/图表/统计”, which can cause the skill to activate for many ordinary conversations not specifically requesting this tool. Over-broad activation increases the chance of unintended routing, unnecessary data exposure to the skill, and user confusion about why the skill was invoked.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The manifest description is entirely in Chinese and presents the skill behavior and trigger phrases in that language only, without indicating that users may choose another language. This can constitute a language/locale policy issue because the skill appears to assume a fixed language without opt-in or justification.

Content

No source excerpt is available for this finding.