Back to skill

Security audit

境外诉讼案例库

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because its reported behavior includes account checks, credential handling, and installing or updating external code beyond the stated legal research purpose.

Consider not installing until you have reviewed the credential setup and external runner source. In particular, avoid sharing an API key in an agent conversation, and verify the runner and its source before allowing installation or updates. The scan was partial, so the evidence does not establish that the artifact is safe.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (10)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

该代码块的实际主要用途是服务可用性和账户状态检查,而不是声明的境外诉讼案例检索与分析。认证和服务检查可以是辅助实现,但此处全部行为均为诊断,其中订阅计费与积分查询属于未声明的账户信息访问,且没有任何公开法律资料检索、诉因归纳或合规分析。首段读取配置时的 '$HOME' 位于 Python 字符串中,不会展开为用户目录,通常会回退到环境变量;后段积分查询则正确展开用户目录并读取配置。该实现细节不改变上述用途不一致的结论。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

An API-key configuration is relevant to this authenticated service, contrary to the mismatch allegation. Nevertheless, the instructions repeatedly require users to paste their secret into the agent conversation and store it using ordinary shell redirection without enforcing restrictive permissions, exposing it to conversation retention and potentially other local users depending on filesystem permissions.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

Installing the repository is related to the documented research runner, so the claimed purpose mismatch is overstated. However, the skill fetches or updates an unpinned repository, potentially from a fallback host, and subsequently executes its Python code with access to the user's files and API credentials; a compromised upstream could therefore cause local code execution.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

文档导出本身可以是研究报告的辅助实现细节,不应单独视为违规能力;但此处提供的技能内容仅包含格式转换命令,没有任何法律资料检索、案例分析或合规启示归纳逻辑。因此,就所提供代码而言,其主要用途与声明存在实质差异。未发现额外的网络访问或敏感资源访问行为。

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 254)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',[]);buddy=[b for s in scenes if s.get('secondary_category')=='涉外法律' for b in s.get('buddies',[])];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 skill mandates Chinese-language reports without offering a language choice.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The output specification explicitly requires a Chinese report, and lines 351–352 repeat this requirement. The skill provides no user opt-in or alternative language selection; its focus on overseas litigation affecting Chinese entities does not establish a need to restrict the report language.

Content

No source excerpt is available for this finding.

The legal-research skill mandates an unrelated referral and social-media promotion workflow.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The low-balance workflow mandates referral, purchase, and social-media promotion prompts and explicitly suppresses other follow-up questions, letting commercial instructions displace the user's research workflow. Its impact is limited because the documented behavior offers choices rather than automatically purchasing credits or publishing a report, but it still creates an unnecessary task-redirection boundary.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 176)May include surrounding context.

Cue API Key:在 cuecue.cn 注册后免费创建,新账号送 500 积分 + 每天 10 积分。点击「创建 API Key」,生成 API Key 后一定要点击「复制」并粘贴发送给您正在使用的智能体。写入 ~/.cue/config.json:

bash
mkdir -p ~/.cue
echo '{"api_key": "sk你的key"}' > ~/.cue/config.json

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 225)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 226)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` |

---