Security audit
Security checks for vulnerabilities and agentic risk
The skill’s Bilibili research and optional downloads fit its stated purpose, with disclosed credential use and third-party package risks.
Your queries and supplied links go to SocialDataX and may consume paid credits. Prefer a reviewed fixed CLI version and a scoped secret store; keep keys out of chat and version control. Choose the download destination explicitly. This review covers the supplied instructions, not the external CLI implementation or future releases.
The search-videos example executes socialdatax-skills@latest, a mutable package reference, with automatic installation approval through npx -y. A compromised future release could execute with the agent's permissions and read its API key or accessible files; this is a CLI supply-chain risk, not an MCP server configuration issue.
The search-articles example automatically executes the package selected by the mutable @latest tag. If a malicious release becomes latest, it could access the agent's environment credentials and files despite the intended read-only research operation.
The content-ID detail command uses npx -y with @latest rather than an immutable reviewed release. A compromised package release could execute arbitrary code with the agent's permissions, including access to SOCIALDATAX_API_KEY.
The URL-based detail example automatically runs socialdatax-skills@latest, allowing the executed code to change without changes to this skill. Package compromise could expose the API key and other resources accessible to the agent; the documented read-only API boundary does not constrain local package code.
The content-ID comments command executes a mutable npm release with npx -y. A malicious future release could run under the agent's identity and steal environment credentials or alter accessible local files, regardless of the comments endpoint being read-only.
The URL-based comments example uses the mutable @latest tag and automatically approves package installation. If package distribution is compromised, this provides a path to code execution with the agent's permissions and access to its API key.
The replies example runs an unpinned package through npx -y. A compromised latest release could execute arbitrary code and access credentials inherited by the CLI; the intended discussion-analysis function does not reduce that local execution capability.
The post-ID reactions command automatically executes whichever package release @latest resolves to. A compromised release could steal the API key or manipulate local resources, even though the intended operation only retrieves reactions.
The URL-based reactions example combines automatic installation approval with a mutable @latest reference. Package compromise could therefore introduce unreviewed code with the agent's privileges and access to runtime secrets.
The user-ID profile lookup runs socialdatax-skills@latest through npx -y, so package updates can change executable behavior without skill review. A malicious release could access the API key and files available to the agent.
The profile-URL lookup automatically executes a mutable npm package release. If the publisher or distribution path is compromised, the executed code could steal inherited credentials or alter local files despite the lookup's read-only purpose.
The user-ID video-list command executes the mutable @latest package with automatic approval. A compromised package release could execute with the agent's permissions, exposing its SocialDataX credential and other accessible resources.
The profile-URL video-list example runs an unpinned release through npx -y. A malicious update could access environment secrets and local files; the documented restrictions on social-account modifications do not sandbox the executable.
The user-ID article-list example automatically executes socialdatax-skills@latest. A compromised future release could run arbitrary code under the agent's identity and expose the API key provided through its environment.
The profile-URL article-list command uses a mutable package tag with npx's automatic approval flag. Package compromise could introduce code that steals credentials or modifies accessible files, independent of the intended read-only API operation.
The user-ID dynamics-list command automatically executes the release referenced by @latest. A malicious release could access the agent's credentials and local resources; no supplied implementation establishes a sandbox that would prevent this.
The profile-URL dynamics-list example runs an unpinned npm release with automatic installation approval. If that release is compromised, it could execute arbitrary code with the agent's permissions and access SOCIALDATAX_API_KEY.
The media-download example automatically executes socialdatax-skills@latest, leaving the executable dependent on a mutable package tag. A compromised release could steal credentials or modify accessible files; requiring an explicit download request does not address package integrity.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
