Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent image-generation workflow for Douyin shop visuals, with disclosed third-party API use, local tool installation, image upload, and credit-spending confirmation requirements.
Install only if you are comfortable using the qhkit/iqinghu service, uploading product images to it, providing an API key, and allowing local CLI setup. Review any prompt before billable generation, and keep international AutoAGC users from overwriting their existing region configuration.
The metadata description says requests for certain image-generation tasks 'must trigger' this skill, using broad mandatory routing language. This can cause the skill to activate in situations where user intent is ambiguous, potentially leading to unsolicited execution paths such as installation, network access, file handling, or third-party API use.
The skill instructs the agent to execute npx @iqinghu/qhkit without pinning a specific version, which allows whatever package version is current at execution time to run. Because npx fetches and executes remote code, a compromised upstream package, malicious new release, or dependency hijack could lead to arbitrary code execution in the agent environment.
This finding duplicates the same persistence behavior: extracting Node into a user-controlled directory and modifying PATH, potentially across sessions. Even though the text includes checksum verification and appears operational rather than malicious, it still introduces persistent state that could affect later tasks and increase attack surface.
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:$PATH"
This finding duplicates the same persistence behavior: extracting Node into a user-controlled directory and modifying PATH, potentially across sessions. Even though the text includes checksum verification and appears operational rather than malicious, it still introduces persistent state that could affect later tasks and increase attack surface.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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 directs the agent to force the domestic cn/prod site configuration and to overwrite existing intl configuration unless the user explicitly says otherwise. This can override an existing account context or routing preference and may send credentials or content to a different service region than the user intended.
The command npx --yes sharp-cli runs an unpinned package from the registry, creating the same remote code execution risk as any transient install-and-run flow. In this skill, it is presented as an automatic fallback for image processing, which increases the chance the agent executes unreviewed third-party code during normal operation.