Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed video-generation helper that uses a qhkit CLI and service token, with setup steps users should approve before running.
Install only if you are comfortable letting the agent install or upgrade Node/qhkit and configure a qhkit service token. Prefer approving those setup steps explicitly, using a scoped token, and avoiding persistent shell changes unless you want qhkit available across sessions.
The manifest says the skill must trigger whenever a user provides a Douyin/TikTok link and asks to '复刻同款、对标、照着做一条', and also lists keywords like '照着做' and '同款视频'. Several of these phrases are broad, common expressions without negative examples or tighter scope constraints, which could cause unintended invocation outside the intended viral-video cloning context.
The skill goes well beyond video-cloning instructions and directs installation of Node.js, global package installation, upgrades, and shell environment modification. Those are system-management actions with broader security impact, increasing the attack surface and giving the skill authority to alter the host environment in ways not strictly necessary for the business task.
The skill describes software installation, upgrades, PATH modification, and credential setup, but does not clearly require an explicit user warning and consent before taking those system-affecting steps. In context, this makes the skill more dangerous because a user asking for video imitation would not reasonably expect environment bootstrapping or persistent config changes.
The skill instructs use of npx @iqinghu/qhkit without a pinned version, which allows execution of whatever package version is current at runtime. That creates supply-chain risk and makes behavior non-reproducible; a compromised or incompatible upstream release could be executed automatically in the user's environment.
This duplicate finding points to the same persistence behavior: extracting binaries into a user directory and exporting a modified PATH. In the context of a content-generation skill, that host modification is broader than necessary and can have lingering effects on later commands and sessions.
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:$PATH"
This duplicate finding points to the same persistence behavior: extracting binaries into a user directory and exporting a modified PATH. In the context of a content-generation skill, that host modification is broader than necessary and can have lingering effects on later commands and sessions.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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 explicitly directs acquisition and persistent configuration of API credentials via qhkit config set --token or environment variables. Handling secrets and writing them into persistent config is security-sensitive and not inherently required for merely analyzing or drafting a video-cloning workflow, creating risk of credential exposure, misuse, or cross-session retention.