Security audit
Security checks for vulnerabilities and agentic risk
The skill has a legitimate AI-text-detection purpose, but it can automatically install or upgrade unpinned external tooling on the user's machine for a simple text-checking task.
Review this skill before installing. It is designed to call a paid external text-detection service and may send the submitted text to that provider. Only allow setup if you are comfortable with npm/qhkit installation, Node installation or upgrade, PATH changes, and use of your Qinghu token; prefer preinstalling and pinning qhkit yourself.
The manifest says the skill must trigger when the user says the listed phrases 'or提出同类需求', which expands activation beyond the concrete examples without defining boundaries. This can cause unintended invocation because many ordinary questions about AI-written text may partially match the intent with no exclusion conditions or negative examples.
The skill’s core purpose is AI-text detection, but it additionally instructs the agent to install Node.js, install npm packages globally, and bootstrap tooling on the host. That is broader than required for the business task and expands the agent from data processing into system modification, increasing the chance of unsafe changes, privilege issues, and execution of unreviewed external code.
The skill includes shell commands, package installation, and environment-changing steps without an upfront warning that the local system may be modified. In an agent setting, missing advance disclosure increases the risk of silent or unexpected host changes, especially when the skill is auto-invoked for a seemingly simple content-analysis task.
The skill instructs fallback execution via npx @iqinghu/qhkit without pinning a specific version, which makes behavior non-deterministic and can pull newly published code at execution time. In an agent context, this increases supply-chain risk because the fetched package version could change unexpectedly or become malicious after the skill was authored.
This duplicate finding points to the same persistence behavior: unpacking binaries under the user home directory and adjusting PATH in a way that may survive the task. Persistent host modification is disproportionate to the skill’s narrow function and increases operational and security risk if the installed tooling later changes or is abused.
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:$PATH"
This duplicate finding points to the same persistence behavior: unpacking binaries under the user home directory and adjusting PATH in a way that may survive the task. Persistent host modification is disproportionate to the skill’s narrow function and increases operational and security risk if the installed tooling later changes or is abused.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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.