Security audit
Security checks for vulnerabilities and agentic risk
The skill’s creator-research workflow fits its stated purpose, with ordinary precautions for third-party queries and software updates.
Send only query terms and profile details you are comfortable sharing with SocialDataX. Set page limits to control credit usage, keep your API key private, and prefer a reviewed, fixed package version over @latest.
The command uses npx -y socialdatax-skills@latest, automatically executing a mutable package version rather than a reviewed release. A compromised package publication could execute with the agent's permissions and access SOCIALDATAX_API_KEY; this is a CLI supply-chain risk, despite the finding's MCP label.
This profile lookup automatically executes the mutable @latest npm release. A malicious or compromised future release could steal the API key or access local resources, regardless of the documented read-only API behavior.
The posts command runs socialdatax-skills@latest without a fixed reviewed version. If the selected release is compromised, its code runs locally with access to the supplied API key and the agent's permitted resources.
This search command automatically executes whichever release the npm @latest tag selects. A compromised publication could execute arbitrary local code; the legitimate creator-search purpose does not constrain package behavior.
The profile command depends on a mutable npm release and uses -y to permit unattended execution. A compromised release could access credentials or local files under the agent's permissions, despite the skill's read-only claim.
The command executes socialdatax-skills@latest rather than a fixed, reviewed release. Package compromise could turn an ordinary content lookup into local code execution with access to SOCIALDATAX_API_KEY.
The series lookup automatically runs code from the mutable @latest release. A compromised release could misuse the agent's local permissions and API credential; read-only business functionality does not prevent that.
This search example uses unattended npx execution against the mutable @latest tag. A malicious package update could execute locally and expose the API key or other resources available to the agent.
The profile command executes an npm version that can change without changes to this skill. If that release is compromised, local code can access the API key and act with the agent's permissions.
The posts command automatically executes the current @latest npm release. A compromised publication could perform actions beyond content retrieval, including credential theft or local file access.
The ID-based lookup still executes socialdatax-skills@latest, so its code is not fixed by the skill's reviewed contents. A compromised release could run with the agent's permissions and access the API credential.
This content-list command uses a mutable package tag with automatic npx execution. Package compromise could enable arbitrary local behavior despite the documented read-only workflow.
The account-info command automatically executes the release selected by @latest. A compromised release could access SOCIALDATAX_API_KEY and local resources under the agent's permissions.
The ID-based posts lookup executes a mutable npm release rather than a fixed reviewed artifact. A malicious update could steal credentials or perform unauthorized local actions.
The series command relies on unattended execution of socialdatax-skills@latest. A compromised publication can execute locally, and the skill's read-only API description does not impose a technical security boundary.
The paginated search example runs a mutable npm release without a version pin. If the package is compromised, execution can expose the API key or affect local resources regardless of pagination settings.
This account lookup automatically executes whatever release @latest resolves to. A compromised publication could act with the agent's local permissions and access its API credential.
The posts lookup executes socialdatax-skills@latest with automatic approval. A malicious future release could execute arbitrary code with access to the API key and resources available to the agent.
The workflow instructs the agent to search using user-supplied keywords and directly query supplied profiles or IDs; the CLI examples also accept complete share text. Although the safety section describes the skill as read-only and discloses API-key use, it does not warn that these inputs are sent to a third-party service, which matters when they contain private research interests or incidental personal information.
No suspicious patterns detected.
