X / Twitter Search

• Outcome: Pass
• Latest audit: May 28, 2026, 4:46 AM
• Version:
  1.0.1

Overview

This looks like a purpose-built X/Twitter search skill with expected third-party API exposure and some invocation/prompt-safety caveats, but no artifact-backed malicious behavior.

Install only if you are comfortable sending X/Twitter search queries to xAI/X services and using the required API credentials. Prefer explicit commands such as searching X/Twitter by name, and treat returned model content as untrusted search output rather than instructions for the agent to follow.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The trigger phrase "find tweets" is broad and likely to match normal user conversation, which can cause unintended invocation of the skill. Because the skill makes external API calls using user-controlled queries, accidental activation could leak prompt content to third-party services or trigger unwanted paid requests.

Shadow Command Trigger

Medium

• Category: Trigger Abuse
• Confidence: 89% confidence
• Finding: The trigger "search x" overlaps with a built-in search command and may shadow or hijack expected platform behavior. That increases the chance that ordinary search requests are routed through this skill, causing unexpected external calls, possible billing, and disclosure of user queries to X/xAI endpoints.

Shadow Command Trigger

Medium

• Category: Trigger Abuse
• Confidence: 89% confidence
• Finding: The trigger "search twitter" similarly conflicts with a generic built-in search capability. In context, this skill sends requests to external services and may incur usage costs, so command shadowing creates a real risk of accidental invocation and unintended data exposure.

Shadow Command Trigger

Medium

• Category: Trigger Abuse
• Confidence: 84% confidence
• Finding: The trigger "find tweets" conflicts with the generic verb "find" and is broad enough to capture casual requests unrelated to this skill. Given that the skill performs outbound API queries, this can lead to unintended third-party disclosure of user text and unnecessary paid API usage.

VirusTotal

64/64 vendors flagged this skill as clean.