ClawHub
Back to skill

Security audit

Devil's Advocate (魔鬼代言人)

Security checks for vulnerabilities and agentic risk

Overview

This Markdown-only skill provides a structured critical-review template and does not request code execution, credentials, network access, or persistence.

Install this if you want direct, structured critique of plans or decisions. Expect a sharp Chinese-language review style and verify any compliance discussion against your actual jurisdiction and professional advice.

A.I.G

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

SkillSpector

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 3)May include surrounding context.

md
---
name: devils-advocate
description: 魔鬼代言人 — 反附和批判性评审 Skill。当用户需要方案评审、决策质疑、批判性反馈、红队思考、避免群体思维时使用。触发场景包括：用户在提出方案/想法/决策后需要外部挑战、想听不同意见、需要多视角审查、做重大决策前想要魔鬼代言人把关。自动召唤5个虚拟利益相关方角色进行尖锐质询，输出结构化改进建议。
---

# Devil's Advocate (魔鬼代言人)

反附和批判性评审 AI Skill。用户输入方案/结论/想法/决策后，自动召唤 5 个虚拟角色从�

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The activation conditions are very broad, covering generic requests for reviews, feedback, challenge, or decision support. In an agent environment, this can cause the skill to trigger on many ordinary conversations and override user intent with an aggressive fixed-format critique, creating prompt-scope overreach and unreliable behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill hard-codes Chinese-language output and China-specific compliance framing without checking the user's locale, language, or jurisdiction. This can mislead users with irrelevant legal guidance and degrade safety by presenting compliance advice as if it were universally applicable.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.