Back to skill

Security audit

Ozon-商品信息追踪 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Ozon product-tracking integration, with the main risk being ordinary but broad setup of qhkit, Node, and a paid API token.

Install only if you trust the Qinghu qhkit package and are comfortable configuring a Qinghu API token and spending Qinghu credits. Prefer a preinstalled or pinned qhkit version in managed environments, and avoid modifying shell startup files unless you explicitly want Node added to your PATH.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The phrase 或提出同类需求时必须触发 makes activation criteria open-ended, allowing the skill to trigger on loosely related requests. In an agent setting, broad trigger language can cause unnecessary tool invocation, unintended paid actions, or collection of data outside the user's precise intent.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill expands its behavior from product tracking into installing Node.js, globally installing npm packages, and modifying the local environment. This materially increases attack surface because a data-query skill should not need to bootstrap general software tooling or perform package-management actions on the host.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The skill instructs use of npx @iqinghu/qhkit without pinning a specific version, which makes execution depend on whatever version is current at runtime. That creates a supply-chain risk and undermines reproducibility, especially because this package is then used to access external services and process user-provided parameters.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
91% confidence
Finding

This finding duplicates the same persistence behavior at the same line: unpacking software into the user's home directory and exporting a modified PATH. In the context of a narrowly scoped data-query skill, that is disproportionate capability escalation and creates durable environmental side effects.

Content

Scanner excerpt · SKILL.md (reported line 93)May include surrounding context.

Linux 二进制安装(装到用户目录,无需 root;先校验官方 SHA256 再解包):

cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | sha256sum -c - mkdir -p "$HOME/.local/lib" && tar -xJf /tmp/node-v22.22.3-linux-x64.tar.xz -C "$HOME/.local/lib" export PATH="$HOME/.local/lib/node-v22.22.3-linux-x64/bin:$PATH"

text

Session Persistence

Medium
Category
Rogue Agent
Confidence
91% confidence
Finding

This finding duplicates the same persistence behavior at the same line: unpacking software into the user's home directory and exporting a modified PATH. In the context of a narrowly scoped data-query skill, that is disproportionate capability escalation and creates durable environmental side effects.

Content

Scanner excerpt · SKILL.md (reported line 93)May include surrounding context.

Linux 二进制安装(装到用户目录,无需 root;先校验官方 SHA256 再解包):

cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | sha256sum -c - mkdir -p "$HOME/.local/lib" && tar -xJf /tmp/node-v22.22.3-linux-x64.tar.xz -C "$HOME/.local/lib" export PATH="$HOME/.local/lib/node-v22.22.3-linux-x64/bin:$PATH"

text

Vague Triggers

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The markdown trigger section contains three short example utterances, but does not define boundaries such as required Ozon context, product-ID specificity, or negative examples. Without constraints, common phrases like '有没有人跟卖' or '评论增长情况' may overlap with broader ecommerce discussions and lead to ambiguous invocation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.