ClawHub
Back to skill

Security audit

The News Tool Free

Security checks across malware telemetry and agentic risk

Overview

This looks like a news lookup skill, but its instructions also route database and storage tasks into the skill without explaining or implementing that scope.

Review before installing. Use it only for public news lookup, and do not send database queries, secrets, private documents, internal URLs, or confidential business data through this skill. The publisher should remove the database/storage wording and clarify external API and callback behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

High
Confidence
97% confidence
Finding
The skill metadata describes a news aggregation tool but also instructs users to use it for database operations, SQL queries, and storage management, which is unrelated to its documented functionality. This kind of scope confusion can cause an agent to invoke the skill in inappropriate contexts and potentially route sensitive data or operational requests to an external news API or shell-based execution flow.

Intent-Code Divergence

High
Confidence
95% confidence
Finding
The capability coverage section claims support for database operations, SQL, and data storage tasks even though the rest of the skill only documents querying a public news API. Overstating capabilities can mislead orchestration logic or users into sending unrelated or sensitive task data through a tool with exec/network access, expanding the risk of data leakage and unsafe tool invocation.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The activation guidance is overly broad and internally conflicting, mixing a news skill with unrelated database and storage scenarios. Ambiguous routing guidance increases the chance that an agent selects this skill for the wrong task, which is particularly risky here because the skill declares exec and external-network-oriented usage patterns.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill accepts query input and an optional callback URL while relying on external public API requests, but it does not clearly warn users that their inputs may be transmitted off-platform. Missing disclosure weakens informed consent and can lead users or agents to include sensitive queries or internal URLs in requests that leave the trusted environment.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.