Security audit
Security checks for vulnerabilities and agentic risk
The skill matches its advertised purpose of generating Wildberries marketplace images, but users should understand it installs external tools and may store a Qinghu API token locally.
Install this only if you are comfortable with qhkit/npm setup, uploading product images to Qinghu AI, and storing a Qinghu API token in local configuration or environment variables. Prefer a sandboxed environment and rotate or remove the token when no longer needed.
The markdown explicitly says '文案用俄语', which imposes a specific language/locale on outputs. Under the policy, forcing a language without user opt-in is a natural-language policy violation unless the constraint is clearly justified and documented as region-specific; here it is phrased as a default requirement rather than a user-selectable option.
The skill's declared purpose is image generation, but it directs the agent to install Node.js, perform global package installation, and alter PATH, which materially broadens its operational scope. That increases the blast radius from a content skill into host modification and toolchain bootstrap, enabling persistence-adjacent changes and making abuse or mistakes more damaging.
The skill instructs use of npx @iqinghu/qhkit without pinning an exact version, which allows whatever package version is current at execution time to be fetched and run. In an agent context this creates a supply-chain risk: a compromised publisher account, malicious update, or breaking release could execute arbitrary code on the host under the agent's privileges.
This duplicated finding points to the same persistence behavior: placing binaries under the home directory and encouraging PATH changes that may outlive the immediate task. In an agent system, such persistence is risky because later tasks may unknowingly execute the newly installed toolchain or depend on altered environment state.
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:$PATH"
This duplicated finding points to the same persistence behavior: placing binaries under the home directory and encouraging PATH changes that may outlive the immediate task. In an agent system, such persistence is risky because later tasks may unknowingly execute the newly installed toolchain or depend on altered environment state.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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 skill instructs the agent to collect a user API key and persist it via qhkit config set or environment variables, which exceeds a narrow image-generation action and introduces credential-handling behavior. Persisting secrets in local configuration can leak across sessions, be reused by other tasks, or be exposed to other local processes depending on host controls.
The fallback image-compression path runs npx --yes sharp-cli without a pinned version, causing live code download and execution from the package registry. Because this is framed as an automatic remediation path, an agent may invoke it without extra scrutiny, expanding exposure to registry compromise or malicious package updates.
The manifest focuses on generating marketplace images via Qinghu AI, but this line authorizes the agent to run local transformation commands and install image-processing dependencies like Pillow or sharp-cli. While intended as a convenience for upload limits, this is an additional local file-processing and package-installation capability not explicitly stated in the skill description.