ClawHub
Back to skill

Security audit

xiaobai-en

Security checks for vulnerabilities and agentic risk

Overview

The skill is a non-executable AI safety messaging framework, but it asks agents to collect and archive youth-oriented social discussion data without clear privacy limits or consent safeguards.

Before installing, require the publisher to limit any social data work to authorized, platform-compliant, anonymized or aggregated data, with clear consent rules, access controls, retention/deletion policy, and explicit protection for minors or youth-related data. The Markdown-only package does not appear to execute code by itself.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill explicitly lists 'data collection of post-2010s thought patterns' as an applicable scenario without any disclosure, consent model, minimization rule, or privacy boundary. Even if the data is sourced from public platforms, profiling young users' opinions and behavior can create privacy, compliance, and misuse risks, especially when the collection purpose is broad and influence-oriented.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The archive plan proposes storing 'Global Gen-Z discussion data' and 'TikTok challenge data' in a repository, but provides no safeguards for consent, anonymization, access control, or retention. Persistently storing opinion and engagement data increases the risk of surveillance-style profiling, unauthorized reuse, and regulatory violations, particularly because the target population may include minors.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.