Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward X search helper that uses the user's xAI API key to send explicit search queries to xAI and return cited results.
Install this only if you intend to use xAI to search X content and are comfortable providing an XAI_API_KEY. Treat search terms as data sent to a third-party API, so avoid sensitive, confidential, regulated, or proprietary queries unless that is acceptable under your policies.
The skill invokes Python, reads an API key from environment/config, and sends user queries over the network, yet it declares no explicit tool scope or permissions boundary. This makes the skill easier to auto-invoke without clear least-privilege constraints or user/operator visibility into its access to secrets and external services.
The description uses broad trigger phrases like finding tweets, searching X/Twitter, and looking up what people are saying, which can match many ordinary user requests. Overbroad routing increases the chance the skill is invoked unexpectedly, causing unnecessary disclosure of user prompts to external APIs and expanding the attack surface.
The usage and notes explain functionality but do not warn that user search terms are transmitted to the external xAI API and used to search X content. Without an explicit disclosure, users may unknowingly send sensitive, proprietary, or regulated information to third-party services.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
def redirect_request(self, req, fp, code, msg, headers, newurl):
raise HTTPError(newurl, code, f"Redirect to {newurl} blocked (auth safety)", headers, fp)
API_URL = "https://api.x.ai/v1/responses"
MODEL = "grok-4.20-reasoning"
TIMEOUT_S = 120
MAX_HANDLES = 10
This code sends the user's search query to the remote xAI Responses API via an HTTP POST request. Although the module docstring says it uses the Grok API, there is no user-facing warning at the point of execution, confirmation prompt, or explicit notice that user-provided input is being transmitted off-system.
No suspicious patterns detected.