Security audit
Security checks for vulnerabilities and agentic risk
The skill does the stated Douyin creator lookup, but it also directs the agent to install and update third-party tooling and make persistent local environment changes.
Install only if you are comfortable letting the agent install or update qhkit/Node tooling and configure a Qinghu token on the machine. Prefer a pre-provisioned or sandboxed environment, confirm paid calls before use, and avoid unpinned npx/@latest execution where reproducibility matters.
The top-level description says requests of similar type 'must trigger', using broad keywords and vague matching. Overly broad mandatory activation can cause the agent to invoke external paid tooling and process URLs or account data when user intent is ambiguous, increasing the chance of unnecessary data disclosure and unwanted actions.
The activation examples are short and underspecified, without scope limits or disambiguation rules. In context, this is more dangerous because the skill also authorizes paid external calls and environment setup, so an ambiguous trigger can cascade into unnecessary external access and cost-incurring actions.
The skill goes beyond data lookup and directs the agent to bootstrap the host by installing Node.js, globally installing packages, modifying configuration, and troubleshooting the local environment. This materially expands the attack surface from a bounded tool invocation into host-level software installation and persistent environment changes, which is risky in an adversarial skill context.
The skill instructs fallback execution via npx @iqinghu/qhkit without pinning an exact package version. That permits whatever version is current at execution time to be fetched and run, weakening reproducibility and increasing supply-chain risk if the package is updated maliciously or incompatibly.
This duplicate finding points to the same persistence behavior: unpacking Node into a user-controlled location and exporting a modified PATH, with instructions to make it survive across commands or sessions. In an agent skill, persistent environment mutation substantially increases risk because it changes the host state outside the immediate task boundary.
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:$PATH"
This duplicate finding points to the same persistence behavior: unpacking Node into a user-controlled location and exporting a modified PATH, with instructions to make it survive across commands or sessions. In an agent skill, persistent environment mutation substantially increases risk because it changes the host state outside the immediate task boundary.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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.