Security audit
Security checks for vulnerabilities and agentic risk
This skill is a read-only SocialDataX helper for Haokan video topic research, with a disclosed API key requirement and no artifact evidence of hidden data access or account-changing actions.
Before installing, make sure you trust SocialDataX and the npm package source because the examples execute the latest published CLI package and use your `SOCIALDATAX_API_KEY`. For stronger control, run it in a limited environment and prefer a pinned package version if available.
The skill instructs agents to execute npx -y socialdatax-skills@latest ..., which fetches and runs the newest package version at execution time instead of a reviewed, immutable release. This creates a supply-chain risk: if the package publisher account is compromised or a malicious/broken version is published, agents will execute attacker-controlled code with access to the environment, including SOCIALDATAX_API_KEY.
This command also uses npx -y socialdatax-skills@latest, causing runtime retrieval and execution of whatever code is currently published under that package name. In an agent context, that is dangerous because the downloaded code may run with user privileges and access secrets or local files, turning documentation into an execution vector.
The skill recommends invoking an unpinned npm package via npx, which is a classic remote code execution supply-chain risk. If a malicious update is published, every subsequent run may execute altered code and exfiltrate the API key or tamper with results.
By relying on npx -y ...@latest, this example makes the skill depend on mutable third-party code at the moment of use. Because the skill is designed to run in environments carrying SOCIALDATAX_API_KEY, a compromised package could steal credentials or execute broader host-level actions.
This finding is a true vulnerability because the command executes the latest published npm package without version pinning, making behavior non-deterministic and vulnerable to upstream compromise. In practice, an attacker controlling the package or release pipeline could run arbitrary code in the agent environment and access the configured API key.
The same unpinned npx pattern appears here, so the trust boundary extends to whatever npm serves at runtime. That is especially risky in agent skills because users may treat these commands as endorsed, while the executed package can change after publication and perform arbitrary actions.
This final command repeats the same supply-chain issue: executing an unpinned npm package at runtime. Even though the skill claims to be read-only in business logic, that does not constrain what the fetched package code can do on the host, so the context does not materially reduce the execution risk.
No suspicious patterns detected.
