Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent Shopee analytics helper, but users should be aware it may install or upgrade the qhkit/Node tooling and store a Qinghu API token before making paid API calls.
Install only if you intend to use Qinghu/qhkit for Shopee site-level analytics. Before letting an agent run setup, confirm any npm/Node installation or upgrade, prefer pinned package versions where possible, verify you are using the correct Qinghu domestic or international account, and approve the paid qhkit calls and expected credit usage.
The description says the skill '必须触发' when the user asks examples like '站点总销售额和客单价' or similar needs. Those phrases are generic and could overlap with ordinary analytics requests that do not clearly specify Shopee or this specific tool, increasing the risk of accidental activation.
The '何时触发' section lists only three short example utterances, including generic requests such as '站点总销售额和客单价' and '对比各站点体量'. Without explicit non-trigger cases or scope boundaries, it is unclear when the skill should not activate, especially for broader ecommerce analysis requests.
The skill goes well beyond describing a site-overview query and includes environment bootstrapping, package installation, token configuration, runtime upgrades, and troubleshooting steps. This materially expands the agent's authority from data retrieval into system modification and credential handling, increasing the attack surface if the skill is auto-invoked or followed blindly.
The skill instructs use of npx @iqinghu/qhkit without pinning a version, which allows whatever package version is current at execution time to be fetched and run. In a security-sensitive agent context, this creates supply-chain and reproducibility risk because behavior can change unexpectedly or a compromised release could be executed.
This duplicate finding points to the same persistent install behavior: extracting binaries into a user directory and exporting a modified PATH. Even with checksum verification, persisting new executables and changing command precedence is dangerous in an agent context because it changes future execution semantics outside the immediate workflow.
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:$PATH"
This duplicate finding points to the same persistent install behavior: extracting binaries into a user directory and exporting a modified PATH. Even with checksum verification, persisting new executables and changing command precedence is dangerous in an agent context because it changes future execution semantics outside the immediate workflow.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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"
The instructions mandate using the domestic Chinese site configuration and tell the agent not to guide users toward the international site unless the user explicitly says they use it. This imposes a locale/account policy choice by default rather than offering the user a choice or checking their preference first.
No suspicious patterns detected.