Back to skill

Security audit

小红书爆款选题分析器

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates the skill benign with high confidence: its completed scans found one limited concern about broad activation terms, while the other assessed dimensions were rated ok.

The scans covered all 8 listed components and completed; the AIG scan also completed without findings. Consider whether broad triggers could interfere with unrelated requests before installing. The benign verdict indicates the scanners found no major concern, but it does not mean risk-free.

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 (1)

The trigger list includes generic terms without requiring a Xiaohongshu-specific context.

Medium
Category
Trigger Abuse
Confidence
96% confidence
Finding

The trigger list explicitly includes generic terms such as “笔记” (notes), “标题” (titles), and “运营” (operations) without requiring Xiaohongshu-related intent, creating a risk of unintended skill activation for unrelated requests. If the host treats these terms as sufficient triggers, the skill's mandatory output requirements could steer responses away from the user's actual task. The impact is limited because the supplied file describes content planning and contains no instructions for sensitive access, code execution, or external transmission.

Content

No source excerpt is available for this finding.