Security audit

X (Twitter) Hot Topics

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent read-only X/Twitter search helper, with some install and scoping caveats users should understand before enabling it.

Install only if you are comfortable providing an X API bearer token and having topic keywords from relevant prompts sent to X/Twitter. Treat the summaries as analysis of returned recent English posts, not verified platform-wide trends, and consider pinning the Tweepy dependency before use in stricter environments.

SkillSpector

By NVIDIA

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 84% confidence
Finding: The documented behavior overstates what the skill actually does: it claims to infer user interests, fetch trending topics, and provide analysis, but the described execution simply submits explicit search queries and returns tweet data. This mismatch can mislead orchestrators or users into trusting the skill with broader autonomous behavior or factual summarization that it does not perform, increasing the risk of misuse, poor policy enforcement, and unreliable outputs.

Context-Inappropriate Capability

Medium

Confidence: 90% confidence
Finding: The wrapper allows the first argument to select any existing .py file in the scripts directory, not just the intended x_hots.py entrypoint. That expands the executable surface of the skill beyond its declared purpose, so a caller can invoke unrelated helper or leftover scripts that may be unsafe, unreviewed, or have different trust assumptions.

Vague Triggers

Medium

Confidence: 76% confidence
Finding: The invocation guidance is broad enough to match common user requests about trends, public opinion, or viral discussions, which can cause the skill to trigger in contexts where users did not specifically ask for X/Twitter retrieval. Over-broad routing can expose user prompts to an external service unnecessarily and increase the chance of irrelevant, biased, or privacy-impacting data access.

Natural-Language Policy Violations

Medium

Confidence: 72% confidence
Finding: Defaulting to English-only retrieval without user consent can silently exclude relevant content and skew outputs, especially for international or multilingual topics. While not a direct code-execution issue, it creates a data-selection bias that can misrepresent public sentiment or omit key perspectives in safety- or policy-relevant summaries.

Unpinned Dependencies

Low

Category: Supply Chain
Content: tweepy>=4.14
Confidence: 96% confidence
Finding: The dependency is only lower-bounded (`tweepy>=4.14`), which allows future unreviewed versions to be installed and can introduce breaking changes or a compromised upstream release into the skill's runtime. In a network-facing skill that retrieves live content from X/Twitter, this increases supply-chain risk because the package will likely be installed automatically in deployment or CI environments.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.