ClawHub
Back to skill

Security audit

JD 撰写台

Security checks for vulnerabilities and agentic risk

Overview

This is a simple JD-writing guidance skill with no code execution, credential access, persistence, or external data flow.

Installers should understand that this skill may activate on broad Chinese recruiting phrases like “职位描述” or “招聘标题”; otherwise it is a low-risk text-generation helper and users should still manually check salary, compliance, and anti-discrimination wording before publishing.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Tp2

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger phrases are broad recruiting terms like '写 JD' and '职位描述' that can match routine user conversation outside a clearly bounded invocation flow. This increases the chance of accidental activation, causing the assistant to switch into this skill unexpectedly and steer the interaction in ways the user did not explicitly request.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The natural-language examples include generic follow-up phrases such as '按上面的结果改成更简短的版本' and '把这些整理成一张表' without tying them to an already active JD-writing session. Such unconstrained examples can teach the orchestrator to invoke the skill on generic editing requests, broadening activation beyond the intended domain.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.