ClawHub
Back to skill

Security audit

xinjianxue-skill-inlaw-cn

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed paid API integration that asks for user consent before binding an account and sending relationship-analysis inputs to the provider.

Install only if you are comfortable linking your Xinjianxue account to the assistant and sending the listed personal details to chinaapi.xinjianxue.com for paid report calls. Provide the one-time authorization code only after verifying the official mini-program flow and the service domain, and do not share screenshots or reuse the code elsewhere.

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 (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill requires the agent to obtain a user's one-time AI authorization code and bind it to an externally hosted service, but the trigger/activation section does not prominently warn at the point of activation that a sensitive credential will later be requested. Even though the body later labels the code as sensitive, delayed disclosure increases the chance that users enter a high-trust flow before understanding they will be asked for a reusable account-binding secret tied to a third-party API.

Static analysis

No suspicious patterns detected.