Back to skill

Security audit

基金值不值得买

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because scanners report a request to share an API key, promotional steps unrelated to fund analysis, and mismatches between the stated purpose and reported behavior; the review is incomplete.

Consider the API key exposure, mandatory promotional workflow, repository setup, and external request before installing. Avoid sharing an API key in chat; configure it locally if you proceed. The scan covered the single listed Markdown component, but analysis was partial because a referenced local path was missing, and the remote repository contents were not assessed.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (8)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

文档导出可以作为基金分析报告的辅助实现细节,本身不应视为违规或无关能力。但所提供代码的全部行为只是转换已有 Markdown 文件,没有调用 Cue、获取基金数据或执行任何所声明的分析,因此其实际主要用途与声明存在实质性差异。此结论仅针对所提供的代码片段。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The supplied chunk performs repository installation or maintenance exclusively. It does not retrieve fund information, calculate metrics, evaluate a manager, or produce a fund assessment. Dependency setup can be a supporting implementation detail, but this chunk contains no invocation or other demonstrated connection to the declared analysis workflow. This assessment is limited to the supplied code and does not assume anything about the downloaded repository's contents.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 311)May include surrounding context.

md
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 for b in s.get('buddies',[]) if b.get('title')=='公募基金全维尽调'];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 workflow asks users to paste an API key into the conversation without warning about credential exposure.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The setup explicitly asks users to paste their API key into the conversation, unnecessarily exposing a credential to chat history, model processing, and potentially logging systems. Although authentication with Cue is legitimate, disclosure of the key could permit unauthorized API usage and consumption of the user's credits.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 286)May include surrounding context.

md
| 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` |

---

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 287)May include surrounding context.

md
| 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 fund-analysis workflow mandates unrelated referral and social-media promotion after successful report delivery (L181–L199).

Low
Category
Excessive Agency
Confidence
94% confidence
Finding

When credits are low, the skill mandates referral, purchase, and social-media promotion and gives that solicitation priority over other follow-up questions. This redirects the assistant beyond the requested fund assessment and can pressure users toward unnecessary purchases or disclosure through report sharing, although it does not automatically purchase or publish anything.

Content

No source excerpt is available for this finding.

Mandatory verbatim Chinese prompts impose an output language without user opt-in.

Low
Category
Output Handling
Confidence
86% confidence
Finding

The low-balance workflow requires the Chinese question and options to be used verbatim and prohibits changes to their wording. This prevents localization to the user's preferred language, and the skill provides neither a language choice nor a justification for this output-language constraint.

Content

No source excerpt is available for this finding.