ClawHub

yanbaoke-research-report-download

Security checks across malware telemetry and agentic risk

Overview

This appears to be an earnings-report helper that uses an FMP API key, with caution needed around API-key storage and any investment advice it may generate.

Install only if you are comfortable providing an FMP API key. Prefer a low-privilege or free-tier key, avoid pasting keys into shared conversations, and store credentials in a safer secret manager or restricted env file instead of a shell startup file when possible. Treat any market or trading output as informational and verify important dates and decisions independently.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

High
Confidence
98% confidence
Finding
The skill explicitly instructs the agent to provide investment analysis and recommendations based on report content. That creates an unsafe behavioral policy because it pushes the agent into regulated, high-stakes financial advice without suitability checks, disclosures, jurisdictional handling, or user-risk assessment, increasing the chance of harmful or noncompliant recommendations.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The instructions tell the user to persist an API key in a shell startup file, which stores the credential in plaintext and causes it to be loaded into every future shell session. This increases exposure through local file disclosure, shell/environment leakage, backups, dotfile syncing, and accidental sharing, without clearly warning the user about those risks or offering safer alternatives.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal