Back to skill

Security audit

王兴思维视角

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only strategy/persona skill with no code execution, credentials, persistence, or data access requirements.

Install this if you want a Chinese-language strategic analysis persona based on public information about Wang Xing. Be aware that broad trigger phrases may invoke it unintentionally, and its supporting notes contain unnecessary internal path references that the publisher should remove.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger list includes generic concepts such as "无限游戏", "长期主义", and "边界思考" without any invocation constraints. These phrases are broad strategic terms that may appear in ordinary business conversation, making activation scope ambiguous and increasing the risk of unintended invocation.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
89% confidence
Finding

The document exposes an internal absolute filesystem path under /home/admin/openclaw/workspace/skills/..., which leaks environment structure and confirms the presence and naming of local skill assets. While not directly exploitable on its own, this kind of enumeration can help an attacker tailor follow-on probing, path-guessing, or social-engineering attacks against the hosting environment.

Content

Scanner excerpt · 反思总结.md (reported line 235)May include surrounding context.

md
|--------|------|------|
| 调研摘要 | ✅ 完成 | `/home/admin/openclaw/workspace/skills/wang-xing-perspective/调研摘要.md` |
| 心智模型列表 | ✅ 完成 | 5 个核心心智模型 |
| 完整 Skill | ✅ 完成 | `/home/admin/openclaw/workspace/skills/wang-xing-perspective/SKILL.md` |
| 验证结果 | ⚠️ 待完成 | 需要实际测试 |
| 反思总结 | ✅ 完成 | 本文档 |

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
86% confidence
Finding

The document discloses an absolute internal filesystem path to the skill file, which is a form of environment and asset enumeration. While not directly exploitable on its own, such path disclosure can help an attacker map the runtime layout, craft more targeted prompts or exploits, and correlate other leaked information about the host environment.

Content

Scanner excerpt · 调研摘要.md (reported line 181)May include surrounding context.

md
## 六、Skill 文件位置

**文件路径**:`/home/admin/openclaw/workspace/skills/wang-xing-perspective/SKILL.md`

**文件大小**:约 6740 字节

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The document specifies that the skill should use Wang Xing's tone and speak in first person, which constrains the generated response style. In this Chinese-language skill context, that is a natural-language behavior constraint presented without any user choice or opt-in, which can be treated as a locale/style policy issue under the stated rule.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

This markdown file is entirely written as a Chinese-only skill research summary and references a downstream SKILL.md artifact, but it does not indicate that language use is optional or user-selectable. Under the policy rule for natural-language violations, forcing a specific language without opt-in can be a locale-policy issue.

Content

No source excerpt is available for this finding.