Back to skill

Security audit

保险产品条款核查

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this insurance policy checking skill suspicious because scanners report API key sharing in chat, a remote fetch, and required referral and purchase prompts; the scan is partial, so the evidence is incomplete.

Consider the suspicious verdict and the scanner's do-not-install recommendation before installing. In particular, do not paste an API key into chat; the scan is partial because a referenced local path was missing, so it does not provide complete evidence about the skill.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
Findings (6)

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 235)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 if s.get('secondary_category')=='保险营销' for b in s.get('buddies',[]) if b.get('title')=='保险产品条款核查'];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 their API key into the conversation without warning about credential exposure.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The setup explicitly asks the user to paste a secret API key into the conversation, unnecessarily exposing it to conversation history and systems that process or retain messages. Although the key is needed for legitimate Cue authentication, collecting it through chat creates avoidable risk of unauthorized API use and credit consumption.

Content

No source excerpt is available for this finding.

Onboarding instructs the agent to solicit a secret API key in chat while assuring users the agent cannot view it.

Medium
Category
Data Exfiltration
Confidence
98% confidence
Finding

The statement that the agent cannot or is not authorized to view the key conflicts with the instruction to paste that same key into the agent's conversation. This can give users a misleading impression of confidentiality and encourage secret disclosure; it is the same underlying credential-handling issue as SQP-2, not a separate compromise path.

Content

No source excerpt is available for this finding.

The insurance-clause checking skill mandates an unrelated referral, purchase, and social-promotion workflow after report delivery (L114–L139).

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The skill makes referral, purchase, and social-promotion prompts mandatory and gives them priority over other follow-up questions, diverting the assistant from insurance analysis into provider promotion. A low-credit warning is relevant, but enforcing promotional wording and encouraging sharing a report-linked identifier exceeds what is needed; no automatic purchase or publication is shown.

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 210)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 211)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` |

---