Security audit
Security checks for vulnerabilities and agentic risk
This read-only Toutiao research skill is coherent, but needs review because it runs an unpinned npm package at runtime.
Install only if you trust the SocialDataX npm package and are comfortable providing `SOCIALDATAX_API_KEY` to a CLI fetched at runtime. Prefer a pinned package version or a preinstalled reviewed CLI before using it in an agent environment.
The manifest description and the full markdown guidance require Chinese comprehension and do not indicate that another language can be used on request. This creates a language/locale policy concern because the skill effectively forces a specific language without user opt-in or documented justification.
The skill instructs execution of npx -y socialdatax-skills@latest, which fetches and runs the latest package version at runtime without pinning to a reviewed release. If the upstream package, dependency chain, or publisher account is compromised, an agent could execute attacker-controlled code in its local environment using available permissions and environment variables such as SOCIALDATAX_API_KEY.
This command again uses npx -y socialdatax-skills@latest, causing the agent to download and execute unpinned remote code at runtime. In the context of an agent skill, that creates a supply-chain execution path where a malicious update could access local files, network resources, or the configured API key.
The unpinned npx invocation here presents the same runtime supply-chain risk: the package version resolved at execution time may differ from the version originally reviewed. Because the skill is intended for automated agent use, a compromised package could turn a read-only research workflow into arbitrary code execution on the host.
This example also relies on npx -y socialdatax-skills@latest, so the executed code is not stable or reviewable over time. An attacker who gains control of the package release process could weaponize the command and exfiltrate secrets or run arbitrary system commands when the agent follows the skill instructions.
No suspicious patterns detected.
