Security audit

News Report

Security checks across malware telemetry and agentic risk

Overview

This looks like a real news-report generator, but it should be reviewed because it sends reports externally and generates unsafe HTML from web/LLM-derived content.

Install only if you are comfortable with search terms and generated reports being sent through Feishu. Review generated HTML locally before sending or sharing it, avoid sensitive topics, and prefer a fixed version that restores TLS verification, escapes/sanitizes HTML fields, and asks for explicit approval before external delivery.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (3)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 90% confidence
Finding: The skill instructs the agent to perform network access, write files to /tmp and a workspace path, and send output externally, but it declares no permissions or trust boundaries. This creates a consent and sandboxing gap: users and the platform cannot accurately assess or constrain what the skill will access, store, or exfiltrate.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 86% confidence
Finding: The documented behavior promises LLM analysis and report delivery, but the actual described implementation appears to scrape search-engine result pages and omits key steps like controlled analysis and sending. This mismatch is dangerous because users may trust the skill with sensitive topics or rely on outputs that are incomplete, misleading, or generated from fragile scraping logic rather than the advertised workflow.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill explicitly says it will send the generated HTML report via Feishu, but the description does not clearly warn users that collected and synthesized data may be transmitted to an external service. This is a real data-handling vulnerability because users may provide sensitive topics, and external transmission without explicit notice or consent can cause confidentiality and compliance issues.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.insecure_tls_verification

HTTPS certificate verification is disabled.

Warn

Code: suspicious.insecure_tls_verification
Location: scripts/fetch_news.py:26