xAI
Security checks across malware telemetry and agentic risk
Overview
This appears to be a straightforward xAI/Grok API wrapper, with expected API-key and data-sharing considerations and minor provenance metadata gaps.
This skill looks coherent for chatting with Grok and using xAI search. Before installing, verify the source because registry provenance is limited, use a dedicated XAI_API_KEY if possible, and avoid passing sensitive prompts or local images/files unless you intend to send them to xAI.
VirusTotal
64/64 vendors flagged this skill as clean.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Prompts, image inputs, and X search queries may leave the local environment and be processed by xAI.
This confirms that user content is sent to the external xAI service. That is expected for this skill, but users should avoid sending sensitive prompts, files, or images unless they are comfortable sharing them with xAI.
Wraps the xAI API (`https://api.x.ai`). Chat uses the standard completions endpoint. X search uses the Responses API with the `x_search` tool for real tweets with citations.
Use the skill only for content you are comfortable sending to xAI, and review xAI's data handling terms for sensitive work.
The API key may allow xAI account usage and billing under the user's account.
The skill uses an xAI API key from the environment or a local config file and sends it as a bearer token to xAI. This is purpose-aligned and bounded to the xAI key, but it is still delegated account authority.
if (process.env.XAI_API_KEY) { return process.env.XAI_API_KEY; } ... const key = config?.skills?.entries?.xai?.apiKey; ... 'Authorization': `Bearer ${apiKey}`Use a dedicated xAI API key if possible, keep it out of shared logs, and monitor xAI usage or billing.
Users have less registry-level assurance that the packaged skill matches the upstream project they expect.
The registry source is not provided. The artifacts do include readable source files and no install-time dependency execution, so this is a provenance note rather than evidence of unsafe behavior.
Source: unknown ... Version: 1.3.1
Verify the listed repository and package contents before installing, especially if using the skill with a real API key.