Description-Behavior Mismatch
- Category
- MCP Tool Poisoning
- Confidence
- 99% confidence
- Finding
文档导出可以作为报告生成的辅助步骤,但所提供代码仅包含格式转换命令,没有客户关系数据检索、产业链分析或拓客建议生成逻辑。因此,就所提供代码而言,其实际主要用途与声明用途存在实质差异。
- Content
Security audit
Security checks for vulnerabilities and agentic risk
ClawScan marks this skill suspicious because scanners report credential exposure, an extra promotional workflow, and a mismatch between its stated prospecting purpose and reported behavior; coverage is partial.
Consider the suspicious verdict before installing. Do not paste a reusable API key into chat; use a local secret-setting method if you choose to proceed. Review whether the reported document conversion and credit-related promotional workflow match what you want, and treat the scan as incomplete because a referenced local path could not be resolved.
文档导出可以作为报告生成的辅助步骤,但所提供代码仅包含格式转换命令,没有客户关系数据检索、产业链分析或拓客建议生成逻辑。因此,就所提供代码而言,其实际主要用途与声明用途存在实质差异。
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
CUE_KEY=$(python3 -c "import json;print(json.load(open('$HOME/.cue/config.json'))['api_key'])" 2>/dev/null || echo "$CUE_API_KEY")
echo "=== 1/4 API Key ===" && [ -n "$CUE_KEY" ] && echo "已配置" || echo "未配置!"
echo "=== 2/4 Cue 服务 ===" && curl -sS --max-time 10 "https://cuecue.cn/api/health" -H "Authorization: Bearer $CUE_KEY"
echo "=== 3/4 搭子 ===" && curl -sS --max-time 10 "https://cuecue.cn/api/playbook" -H "Authorization: Bearer $CUE_KEY" | python3 -c "import sys,json;scenes=json.load(sys.stdin).get('data',{}).get('scenes',[]);buddy=[b for s in scenes if s.get('secondary_category')=='商机挖掘' for b in s.get('buddies',[]) if b.get('title')=='产业链上下游拓客'];print(f'可用:{len(buddy)}个') if buddy else print('暂不可用')"
echo "=== 4/4 剩余积分 ===" && python3 -c "
import json, urllib.request, os
p = os.path.expanduser('~/.cue/config.json')
The setup repeatedly asks users to paste an API key into the conversation, unnecessarily exposing a reusable credential to chat history, logging, and any systems with access to those records. Although authentication to Cue is necessary, conversational disclosure is not and could enable unauthorized API usage or credit consumption within the key's permissions.
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
| 目标格式 | 依赖 | macOS | Ubuntu |
|----------|------|-------|--------|
| Word (.docx) | pandoc | `brew install pandoc` | `sudo apt install pandoc` |
| PDF (.pdf) | pandoc + LaTeX | `brew install --cask basictex` | `sudo apt install texlive-xetex` |
---
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
| 目标格式 | 依赖 | macOS | Ubuntu |
|----------|------|-------|--------|
| Word (.docx) | pandoc | `brew install pandoc` | `sudo apt install pandoc` |
| PDF (.pdf) | pandoc + LaTeX | `brew install --cask basictex` | `sudo apt install texlive-xetex` |
---
The skill mandates a vendor-promotional interaction and suppresses other follow-up questions when credits are low, redirecting the assistant beyond the user's prospecting request. This is a limited workflow-hijacking risk rather than demonstrated fraud or exfiltration: the options include declining, and no purchase or social posting is automatically executed.
The skill requires the supplied Chinese question and options to be used verbatim, explicitly prohibiting changes to their wording. This fixes the language of that user-facing interaction without user opt-in or a documented locale-specific justification.