ClawHub
Back to skill

Security audit

stock-market-hotspot-discovery

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent market-hotspot tool that sends the user's query to East Money and can save the returned Markdown report locally.

Install only if you are comfortable sending market-hotspot questions to East Money using an EM_API_KEY you can rotate or revoke. Use --no-save if you do not want returned reports written to disk, and avoid including confidential trading plans or personal data in queries.

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 (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill declares only an environment-variable requirement in metadata, but the documented behavior clearly includes outbound network access to a third-party API and optional local file writes. This mismatch weakens policy enforcement and review because operators may approve the skill without understanding its full capabilities, increasing the chance of unintended data egress or filesystem side effects.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill writes generated market reports containing user-supplied query context and remote API output to a predictable local directory by default. Persistent local storage can leak sensitive prompts, analysis history, or regulated/business-confidential research to other local users, processes, backups, or later tooling, especially since the manifest suggests a returned report rather than durable storage.

Description-Behavior Mismatch

Low
Confidence
89% confidence
Finding
The implementation depends on a third-party Eastmoney network service that is not disclosed in the manifest, creating an unexpected data-sharing boundary. Even if the service is legitimate, undisclosed transmission of user queries can violate privacy expectations, enterprise policy, or compliance requirements.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
User question content is posted to a remote API without an in-context warning or consent mechanism, so users may unknowingly send sensitive investment research, proprietary strategy, or personal data off-host. In a financial-analysis skill, this context raises risk because prompts may contain nonpublic trading rationale or commercially sensitive information.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.