stock sector hunter

• Outcome: Review
• Latest audit: May 28, 2026 · 7:48 AM UTC
• Version: 1.0.0

Overview

The main stock-search function is disclosed, but the package also includes undocumented stock-screening scripts and broader execution behavior that users should review before installing.

Review before installing. This does not show credential theft, destructive actions, or account mutation, but it performs networked financial lookups, depends on a search API credential, includes undocumented bulk stock-screening scripts, uses shell-based subprocess execution, and writes one temporary output file. Treat its market output as informational only, not investment advice.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (9)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

cmd = f'python3 "{script_path}" "{query}" --count {count}'
    
    try:
        result = subprocess.run(
            cmd,
            shell=True,
            capture_output=True,

• Confidence: 97% confidence
• Finding: result = subprocess.run( cmd, shell=True, capture_output=True, text=True, timeout=30, cwd="/root/.openclaw/workspace/ski

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 82% confidence
• Finding: The skill advertises networked stock analysis, but the static finding indicates additional code capabilities including file_write and shell without any declared permissions. Undeclared privileged capabilities reduce transparency and make it harder for a host agent or reviewer to enforce least privilege, especially in a skill that also depends on external search and Python execution.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence
• Finding: This is a significant description-behavior mismatch: the skill claims to perform web-search-based sector discovery, but analysis indicates hidden bulk stock scanning, technical-indicator screening, concurrent API harvesting, and writing results to /tmp/ma_stocks.txt. Hidden functionality is dangerous because users and orchestrators may approve a low-risk search skill while it actually performs broader data collection and local file output beyond the declared purpose.

Description-Behavior Mismatch

High

• Confidence: 97% confidence
• Finding: The implementation materially diverges from the advertised skill behavior: instead of finding active sectors and limit-up leader stocks, it performs a moving-average stock scan over individual securities. In an agent setting, this is dangerous because downstream users or automations may trust the manifest and act on mismatched outputs, leading to incorrect financial decisions and unreliable tool behavior.

Intent-Code Divergence

Medium

• Confidence: 91% confidence
• Finding: The module docstring explicitly describes a moving-average stock selector, which contradicts the skill's declared purpose of identifying strong sectors and limit-up leaders. This inconsistency increases the risk of operator confusion, misrouting of requests, and silent misuse of the tool in financial contexts where accuracy of function is important.

Description-Behavior Mismatch

High

• Confidence: 97% confidence
• Finding: The implementation materially diverges from the skill manifest: instead of identifying hot sectors and limit-up leader stocks, it performs a moving-average stock screen over individual equities. In an agent ecosystem, this is dangerous because callers may invoke the skill expecting one behavior while receiving unrelated market-selection output, causing unsafe automation, misleading financial decisions, and loss of trust in skill-to-manifest binding.

Intent-Code Divergence

Medium

• Confidence: 94% confidence
• Finding: The module docstring and user-facing banners explicitly present the tool as a '均线多头选股器', directly contradicting the manifest's description of a hot-sector/leader-stock detector. This inconsistency increases the likelihood of operator confusion, incorrect invocation, and downstream systems trusting mislabeled functionality in financial contexts where semantic accuracy matters.

Vague Triggers

Medium

• Confidence: 72% confidence
• Finding: Broad trigger phrases like “今日热门” and “热点板块” can cause unintended invocation when a user is discussing markets generally rather than explicitly requesting this skill. Accidental activation is a security and safety concern because it may invoke network access and any hidden side effects tied to the skill without clear user intent.

Vague Triggers

Medium

• Confidence: 74% confidence
• Finding: Examples such as “今日涨停股” and “帮我探测强势板块个股” are generic enough to overlap with normal conversation or other finance tools, increasing the chance of unintended routing. In the context of a skill with network and possibly undeclared extra behaviors, ambiguous activation increases the risk surface.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.