ClawHub
Back to skill

Security audit

Stock Filter Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is a stock-screening helper, but it asks for command execution and file-writing authority while its free-version behavior and routing scope are unclear.

Review before installing. The skill does not show malicious behavior, credential theft, destructive commands, or hidden exfiltration, but it should be narrowed before normal use: remove or clearly scope write access, resolve the export contradiction, define exactly what scripts may be run, and limit activation to stock-screening tasks.

SkillSpector

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

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill declares `write` and `exec` capabilities while the free-edition documentation says export/output writing is unsupported, creating a permission-to-behavior mismatch. This is dangerous because an agent may still grant file-writing or export-like actions based on the manifest, enabling unintended data exfiltration, local file modification, or broader side effects than users expect.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The documentation says the free version does not support export, yet elsewhere says core execution supports create/query/export operations. Contradictory security-relevant behavior descriptions can cause agents or operators to authorize unsafe actions under false assumptions, especially when combined with `write` and `exec` privileges.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The skill advertises very broad triggers such as generic data analysis, report generation, statistics, and visualization, far beyond a narrowly scoped stock screener. Overbroad activation increases the chance the skill is invoked for unrelated tasks, unnecessarily exposing `exec` and `write` capabilities in contexts where they are not needed.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The capability coverage section lists vague keyword fragments like 'Use', 'when', '数据分析', and '自动化工作流场景', which function as catch-all triggers rather than precise boundaries. In skill-routing systems, this can lead to inappropriate selection and execution of a privileged skill, increasing attack surface and the likelihood of misuse.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.