Back to skill

Security audit

Deepthink Expert

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only expert workflow skill that changes analysis style but does not install code, fetch payloads, or request credentials or persistence.

Install this only if you want a Chinese-language expert-analysis workflow that may use more tokens and more structured reasoning. For legal, medical, compliance, security, investment, or large financial decisions, require explicit confirmation and independent professional review before acting on its conclusions.

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
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 4)May include surrounding context.

md
---
name: deepthink-expert
version: 2.0.0
description: "升级版Expert Mode | 集成Claude Code多Agent并行机制 | 代码审查(4Agent并行+置信度评分) | 特性开发(7阶段工作流) | 复杂决策(7专家会诊+交叉验证) | 自动Ralph循环调试"
author:
  name: Hao Zheng
---

# 🧠 DeepThink Expert Mode v2.0

## 用途
处理复杂/高难度任务时自动切换为**多Agent专家协作模式**,集成Claude Code源码泄露中的核心技术。

---

## 自动触发机制

### 🔴 强制触发
- 用户说"专家模式"、"多Agent"、"并行分析"
- 涉及大额(>$100)/法律/合规/医学/安全决策
- �

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill defines broad automatic activation criteria such as legal, medical, compliance, security, and even user uncertainty, which can cause the mode to engage without clear user consent or task suitability. In an agent skill, this increases the chance of overreach, unnecessary delegation patterns, or unsafe workflow changes for sensitive decisions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description and all user-facing instructions are written entirely in Chinese, and example invocations and outputs assume Chinese interaction. There is no indication that users may choose another language or locale, which constitutes a language-policy concern when no opt-in is provided.

Content

No source excerpt is available for this finding.