Back to skill

Security audit

Item manager

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Chinese item-location helper that stores and retrieves user-provided item records, with broad natural-language triggers users should understand before installing.

Install this only if you want a Chinese-language assistant to remember item locations and expiry details. Avoid sharing sensitive item/location details unless you are comfortable storing them, and review records if casual Chinese statements might be interpreted as inventory updates.

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 (3)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger list contains extremely common Chinese words such as “在”, “放”, “哪里”, and “列出”, which are likely to appear in ordinary conversation unrelated to item management. This can cause accidental invocation, unintended access to memory/filesystem-backed item records, and over-collection or disclosure of user data because the skill may activate when the user did not intend to use it.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The README states that natural phrases like “xx 在 xx” are recorded directly and that no explicit “remember” prefix is required. This creates an overly broad trigger surface where ordinary conversation can be misinterpreted as a storage-update command, leading to unintended state changes, poisoned inventory data, or prompt-injection-style social engineering through casual text.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The title, description, triggers, usage examples, and instructions are entirely in Chinese, and the skill says users can interact via normal speech without indicating any multilingual support or opt-in. This effectively imposes a language constraint in natural-language behavior without stating that the skill is intentionally region-specific or allowing the user to choose a language.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.