Back to skill

Security audit

Ozon-关键词详情 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

The skill is narrowly aimed at Ozon keyword lookup, but it also tells the agent to install and update local runtime tooling and configure credentials during normal use.

Install only if you are comfortable letting this skill install or update Node/qhkit tooling and store Qinghu API configuration on the machine. Prefer preinstalling a reviewed, pinned qhkit version yourself, then use the skill only after confirming paid credit usage and the correct Qinghu account region.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (8)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest says the skill must trigger when the user gives the listed examples "or提出同类需求时必须触发", which broadens activation beyond explicit phrases. Without scope limits or negative examples, this creates ambiguity about what counts as a similar request and increases the chance of unintended invocation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The "何时触发" section lists only two example utterances, including a very general phrase like "这个词怎么样", but does not define required constraints such as needing Ozon keyword-detail intent, keyword IDs, or exclusion cases. This ambiguity makes activation criteria unclear and may overlap with ordinary analysis requests.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill claims to perform a narrow Ozon keyword query, but it also directs host-level bootstrap steps including installing Node, installing npm packages, setting config, and modifying environment state. This violates least privilege and expands the attack surface far beyond the business function of reading keyword metrics.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

These instructions authorize host-level software installation and configuration changes that are not justified by the narrow query purpose of the skill. Even if intended for usability, such behavior enables arbitrary code introduction, configuration drift, and persistence on the host system.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The skill instructs use of npx @iqinghu/qhkit without pinning an exact version, which can fetch and execute whatever package version is current at runtime. In a skill whose stated purpose is keyword-detail lookup, this creates unnecessary supply-chain risk and makes behavior non-reproducible across sessions.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
97% confidence
Finding

This duplicated finding identifies the same persistence behavior: unpacking a runtime into the user's home directory and altering PATH. Such host-state changes outlive the task and are disproportionate to the skill's narrowly stated purpose, making the context more dangerous rather than less.

Content

Scanner excerpt · SKILL.md (reported line 91)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
97% confidence
Finding

This duplicated finding identifies the same persistence behavior: unpacking a runtime into the user's home directory and altering PATH. Such host-state changes outlive the task and are disproportionate to the skill's narrowly stated purpose, making the context more dangerous rather than less.

Content

Scanner excerpt · SKILL.md (reported line 91)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

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file states the skill "对接国内版 iqinghu.com" and instructs agents to directly send domestic-site key guidance and not use international-site responses unless the user explicitly says they use the international account. This imposes a locale/site preference by default rather than offering the user a choice, which is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.