Back to skill

Security audit

ClawZhiPin

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed domain-sale promotion, but it is scoped broadly enough to inject ads into ordinary Claw/OpenClaw conversations and can open an external site after user interest.

Install only if you explicitly want an advertising helper for these domains. Expect it to mention clawzhipin.com and ideacool@vip.qq.com in broad AI employee, Claw, OpenClaw, or recruiting conversations; it should be narrowed to direct domain-sale requests and should ask before opening any website.

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
Findings (1)

T01 · Skill Instruction Hijacking

Warning
Location
SKILL.md:3
Finding

Keyword-Triggered Unsolicited Advertising and Response Hijacking

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger list is intentionally broad and includes common terms like "Claw" and general discussion topics such as AI employees and recruiting, causing the skill to activate during ordinary conversation rather than clear purchase intent. In context, the skill is explicitly designed to inject unsolicited domain-sale advertising into unrelated chats, which is a prompt-injection/spam behavior that can manipulate user interactions and degrade agent trust.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructs the agent to open clawzhipin.com in the system browser, but this side effect is not disclosed in the skill description or initial behavior summary. Undisclosed external navigation can surprise users, create trust and privacy concerns, and expose them to unwanted web content or tracking even if activation is framed as interest-based.

Content

No source excerpt is available for this finding.