Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed LinkPix/qhkit integration for product image background replacement, with ordinary dependency and account-token setup risks.
Install only if you are comfortable using the qhkit/LinkPix service, providing its API token, spending account credits, and sending selected image files to that provider. Prefer a pinned or reviewed qhkit version in managed environments.
The description says the skill 'must trigger' whenever the user asks to change an image background or place a product into a scene, and lists broad phrases like '换背景' and '更换背景'. These phrases are common requests that may overlap with other image-editing tools, and the file does not provide exclusion conditions or negative examples to distinguish when this skill should not activate.
The skill instructs fallback execution via npx @iqinghu/qhkit without pinning an exact version. That allows whatever the registry serves at invocation time to be fetched and executed, creating a supply-chain risk if a malicious or breaking release is published or if dependency resolution is tampered with.
This duplicate finding identifies the same behavior: unpacking Node into a user-controlled persistent directory and exporting it into PATH. In the context of an agent skill, instructions that bootstrap and persist developer tooling are more dangerous because they normalize environment mutation and make future command execution depend on newly introduced binaries.
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:$PATH"
This duplicate finding identifies the same behavior: unpacking Node into a user-controlled persistent directory and exporting it into PATH. In the context of an agent skill, instructions that bootstrap and persist developer tooling are more dangerous because they normalize environment mutation and make future command execution depend on newly introduced binaries.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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 require using domestic-site key-retrieval wording and state that CLI messages are already user-facing Chinese, implying a fixed Chinese-language interaction model. There is no indication that the user can choose another language or locale, nor a clear justification that the skill is strictly limited to Chinese-speaking users.