Back to skill

Security audit

目标信息网络搜索及总结

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed web research workflow for finding and summarizing hard-to-locate public information, with no code execution, persistence, or credential access.

Before installing, consider that the skill is written in Chinese and is designed for expansive web research, including translated/local-language queries and third-party source checks. Use it for public-source research tasks, and make sure any sensitive competitive or regulatory research stays within legal and ethical boundaries.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file’s natural-language description and operating instructions are entirely in Chinese, and there is no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking context. Under the language/locale policy rule, this is a natural-language policy concern because it imposes a locale without opt-in or justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The methodology instructs the agent to search in the target region’s native language whenever information is region-specific, without mentioning user awareness, consent, or safeguards. This can broaden data collection and query formulation in ways the user did not explicitly request, increasing the risk of over-collection, opaque search behavior, and potentially privacy- or policy-sensitive retrieval in cross-border contexts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.