Investment Research Analyst

• Outcome: Review
• Latest audit: May 29, 2026 at 9:26 PM UTC
• Version:
  1.0.0

Overview

This investment research skill is mostly coherent, but it can generate and deploy a dashboard without clearly asking the user first.

Install only if you want a Chinese-oriented stock research workflow that may create deployable dashboards. Before using it, require the agent to ask before writing files or deploying anything, confirm whether the deployment is public or private, and avoid entering private portfolio, client, account, or proprietary research information unless the destination is controlled.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger list is broad enough to activate on generic investing or company-research requests, which can cause the agent to enter a specialized workflow unexpectedly. In this skill, that increases the chance of unrequested financial-analysis behavior and downstream actions such as report generation or deployment steps being invoked without clear user intent.

Natural-Language Policy Violations

Medium

• Confidence: 87% confidence
• Finding: Forcing Chinese-language operation/output without checking user preference can mislead users, degrade comprehension, and cause them to approve or rely on content they cannot fully review. While not directly enabling code execution, it weakens informed consent and is riskier here because the skill produces financial analysis that users may act upon.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The workflow includes creating and deploying an interactive dashboard but does not require explicit user consent or warn that content may be published externally. In this context, the dashboard could contain generated financial conclusions, scraped data summaries, or sensitive research artifacts, making silent deployment an avoidable exposure risk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The deployment steps instruct saving HTML to the project directory and using a deploy tool, but they omit warnings about filesystem side effects and external publication. This is dangerous because it normalizes writing artifacts and pushing them to a hosted endpoint without transparency, which can expose proprietary analysis or create unintended public content.

VirusTotal

63/63 vendors flagged this skill as clean.