Back to skill

Security audit

Shopee-店铺类目分布 | 青虎AI

Security checks for vulnerabilities and agentic risk

Overview

This Shopee analytics skill is coherent, but it also tells the agent to install runtimes and CLI tools from the network and configure a service token during normal use.

Install only if you are comfortable with qhkit accessing Qinghu/Shopee analytics data and with the agent potentially installing Node/qhkit tooling and configuring a qhkit token. Prefer preinstalling a vetted qhkit version yourself and avoid the unpinned npx or ad hoc Node download path during normal use.

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 (7)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description says the skill must trigger when the user asks the listed examples 'or提出同类需求', which broadens activation beyond the explicit phrases and leaves the boundary of matching unspecified. This can cause unintended invocation because similar everyday analytical requests may overlap without negative examples or exclusion conditions.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This is a query-oriented analytics skill, but it contains extensive instructions to install software, configure tokens, and modify the runtime environment. Expanding from data lookup into environment bootstrapping materially increases attack surface because the agent may perform privileged setup actions unrelated to the user’s immediate task.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The skill instructs use of npx @iqinghu/qhkit without pinning an exact version or integrity mechanism. That can cause execution of an unexpected package version at runtime and enlarges supply-chain risk, especially because the same skill also tells the agent to install and execute tooling from the network.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill directs downloading and installing Node.js from external URLs, including alternate mirror infrastructure, then using it to execute subsequent commands. Even with checksum guidance, this exceeds the declared capability and creates unnecessary remote code execution and supply-chain exposure inside the host environment.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
94% confidence
Finding

This duplicate finding reflects the same persistence behavior: installing Node under the user's home directory and exporting a new PATH. In the context of a query skill, that is an unnecessary and dangerous state change that can influence future commands and broaden compromise impact if the downloaded toolchain is malicious or tampered with.

Content

Scanner excerpt · SKILL.md (reported line 96)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
94% confidence
Finding

This duplicate finding reflects the same persistence behavior: installing Node under the user's home directory and exporting a new PATH. In the context of a query skill, that is an unnecessary and dangerous state change that can influence future commands and broaden compromise impact if the downloaded toolchain is malicious or tampered with.

Content

Scanner excerpt · SKILL.md (reported line 96)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
88% confidence
Finding

The skill strongly steers users toward a specific domestic account and token flow unless they explicitly opt out. That can misdirect credential handling, create account confusion, and increase the chance that users enter secrets into the wrong service context, especially when paired with CLI configuration instructions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.