Security audit
Security checks for vulnerabilities and agentic risk
The skill has a coherent Douyin-comment collection purpose, but it asks the agent to install and update external tooling and modify local qhkit credentials/configuration, which needs review before use.
Install only if you are comfortable with the agent installing or updating Node/qhkit and changing qhkit API-key configuration on your machine. Before use, require explicit confirmation for installs, upgrades, token changes, and every paid data-collection batch, and avoid invoking it unless the request clearly names Douyin video comments or includes a Douyin video URL.
The requirement that similar requests 必须触发 is ambiguous and creates pressure to invoke the skill even when the match is fuzzy. In an agent setting, ambiguous mandatory triggering can bypass normal intent disambiguation and lead to unintended external calls or charges.
The requirement that similar requests 必须触发 is ambiguous and creates pressure to invoke the skill even when the match is fuzzy. In an agent setting, ambiguous mandatory triggering can bypass normal intent disambiguation and lead to unintended external calls or charges.
The trigger list again includes a generic phrase without sufficient context, reinforcing a broad auto-invocation policy for requests that may not actually be about Douyin comments. Because this skill can lead to external calls and billing, weak trigger specificity is materially risky rather than merely a UX issue.
The skill goes far beyond comment retrieval by instructing the agent to install Node, install packages, download binaries, modify PATH, and bootstrap an execution environment. This materially expands the attack surface and can cause the agent to make persistent system changes or execute untrusted tooling unrelated to the narrow user task.
The skill directs fallback execution via npx @iqinghu/qhkit without pinning an exact version, which makes behavior dependent on whatever package version is resolved at runtime. In a security-sensitive agent context, this increases supply-chain and reproducibility risk because a newer or compromised package version could be fetched and executed unexpectedly.
This duplicate finding points to the same persistence behavior: unpacking software into the home directory and altering execution paths. In the context of a simple comment-collection skill, such persistence is unjustified and increases the risk of unintended system modification and future execution of newly installed binaries.
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:$PATH"
This duplicate finding points to the same persistence behavior: unpacking software into the home directory and altering execution paths. In the context of a simple comment-collection skill, such persistence is unjustified and increases the risk of unintended system modification and future execution of newly installed binaries.
cd /tmp && curl -fsSLO https://nodejs.org/dist/v22.22.3/node-v22.22.3-linux-x64.tar.xz cd /tmp && curl -fsSL https://nodejs.org/dist/v22.22.3/SHASUMS256.txt | grep ' node-v22.22.3-linux-x64.tar.xz$' | 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 reading configuration state and altering credentials or environment (qhkit config show, config set, QHKIT_TOKEN, QHKIT_ENV) even though the stated purpose is only collecting Douyin comments. Accessing and modifying local auth/config state can expose secrets, overwrite existing configuration, or interfere with unrelated accounts and services.
The file instructs the agent to send specific Chinese guidance text to the user, including domestic-site key retrieval wording, without offering a language choice. This can violate language/locale policy when the user prefers another language or has not opted into Chinese-only communication.
No suspicious patterns detected.