ClawHub
Back to skill

Security audit

stock-terminal

Security checks across malware telemetry and agentic risk

Overview

This is a read-only stock research skill that clearly discloses its API key use, optional CLI path, and bounded news-title fetching guidance.

Install this only if you are comfortable providing a SentiSense API key for read-only market data. Prefer environment-variable authentication unless you specifically want the CLI credential cache, and implement news headline fetching only with the hardened restrictions described in the skill.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill explicitly encourages shell execution via `npx` and offers local credential persistence under `~/.config/sentisense/`. Even though the package/version is pinned and the skill is read-only in business purpose, invoking external code and storing secrets on disk expands the attack surface beyond simple API use and can expose hosts that allow shell access.

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill expands from stock-data retrieval into fetching arbitrary external URLs from article feeds for headline and embed resolution. Although it includes substantial safety guidance, this still introduces SSRF/open-fetch risk, redirect abuse, and untrusted third-party content handling that exceeds the declared read-only stock-terminal scope.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.