ClawHub
Back to skill

Security audit

china-news-tool-free

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a straightforward RSS news aggregation helper; its network use and optional local file writes are visible and fit the stated purpose.

Install only if you are comfortable with the agent making HTTP requests to the listed news RSS providers and writing generated briefs or cache files when you run those examples. Review the commands before execution, especially pip install, generated script creation, and output directory settings.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill declares the exec tool even though its stated purpose is simple RSS fetching, classification, and brief generation. Granting arbitrary command execution broadens the attack surface substantially: a prompt or modified workflow could execute shell commands, write files, install packages, or access local environment data far beyond what is needed for this task.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill performs outbound requests to third-party RSS endpoints but does not clearly warn users in the primary description that external network access will occur. In an agent setting, silent network access can surprise users, leak usage metadata such as IP/user-agent/timing, and violate expected isolation or compliance requirements.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill documents saving generated briefs to disk and later includes file-based caching examples, but does not clearly warn users that local files/directories may be created. In agent environments this can create unintended persistence, overwrite concerns, or retention of potentially sensitive derived outputs without the user's informed consent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.