Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent Ozon brand-data helper, with disclosed paid API use and setup steps, but users should be aware it may install/configure qhkit and store an API token.
Install only if you are comfortable using Qinghu/qhkit for Ozon data, providing a Qinghu API token, and allowing the agent to install or update local Node/npm tooling. Confirm credit costs before paid calls and prefer a pinned or preinstalled qhkit version in controlled environments.
The description states the skill must trigger when the user says the listed examples “or提出同类需求”, but it does not define what counts as a similar request. This lacks specificity on trigger scope and can overlap with many adjacent ecommerce-analysis requests.
The trigger example “几个品牌对比” is broad everyday language and, by itself, does not clearly limit activation to Ozon brand-detail queries. Without negative examples or tighter scope, it could cause unintended invocation for generic brand-comparison requests outside this skill’s domain.
The skill goes far beyond its stated purpose of querying brand details and includes extensive environment bootstrapping, package installation, Node setup, token configuration, and troubleshooting instructions. This expands the attack surface by encouraging host modification and secret handling for a routine data query, which is dangerous in an agent setting where the skill content is trusted operational guidance.
The skill instructs use of npx @iqinghu/qhkit without pinning an exact version, which can fetch and execute the latest package at runtime. In a skill context, this creates a supply-chain risk and makes executions non-reproducible, especially because the same document also encourages installing and running tooling on the host environment.
This duplicate finding flags the same persistence behavior: extracting Node into the user's home directory and exporting a modified PATH. In context, the danger is not the specific binary alone but that the skill normalizes persistent environment mutation unrelated to the narrow business function of brand-detail lookup.
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:$PATH"
This duplicate finding flags the same persistence behavior: extracting Node into the user's home directory and exporting a modified PATH. In context, the danger is not the specific binary alone but that the skill normalizes persistent environment mutation unrelated to the narrow business function of brand-detail lookup.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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"
No suspicious patterns detected.