A 股多空辩论

• Outcome: Review
• Latest audit: May 29, 2026 at 9:27 PM UTC
• Version:
  2.1.0

Overview

This stock-analysis skill is mostly aligned with its purpose, but it asks the agent to run an unverified proxy binary from /tmp and stop processes by broad name matching.

Review before installing. Only use this skill if you trust and can verify the mihomo binary and config it expects under /tmp, are comfortable sending stock data and prompts to BigModel GLM, and can tolerate a local proxy process being started and stopped. Avoid shared servers unless proxy handling is changed to a verified binary, explicit user approval, exact PID tracking, and reliable cleanup.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Rogue Agent: Self-Modification, Session Persistence
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Session Persistence

Medium

• Category: Rogue Agent

Content

# 启动代理（如未运行）
stock_start_proxy() {
    pgrep -f mihomo_bin > /dev/null && return
    nohup /tmp/mihomo_bin -f /tmp/mihomo_config.yaml -d /tmp/mihomo_data > /tmp/mihomo.log 2>&1 &
    sleep 2
    curl -s --connect-timeout 5 --proxy ${STOCK_PROXY:-http://127.0.0.1:7890} https://httpbin.org/ip | grep -q origin || echo "PROXY_DOWN"
}

• Confidence: 93% confidence
• Finding: nohup

VirusTotal

65/65 vendors flagged this skill as clean.