Back to skill

Security audit

AI Task Companion

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese-language AI tool recommendation skill with no bundled code or persistence, but users should review external tool actions before execution.

Install only if you want a Chinese-language assistant that recommends and coordinates AI tools. Before agreeing to execution, check which external tool will be used, what content will be sent, whether browser automation or scraping is involved, and whether there may be account, privacy, cost, or website-policy effects.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger phrases are extremely broad and overlap with normal conversational requests such as '帮我做...' and '我需要...'. This can cause the skill to activate unintentionally in many unrelated contexts, steering users into tool execution or recommendations they did not explicitly request, which increases the risk of unauthorized actions and accidental data sharing with downstream tools.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest description and the rest of the skill content are written in Chinese, and the file does not indicate that this is a China-specific or Chinese-only skill nor offer any user language choice. This can violate a language/locale policy when a skill implicitly forces a single language without opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly offers to 'directly help complete' tasks, but it does not warn users about privacy implications, data transfer to external tools, possible costs, or side effects of execution. In a skill designed to broker multiple AI tools, this omission can lead users to unknowingly expose sensitive content or approve actions with external effects.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The tool reference list includes 'agent-browser' for web automation and data scraping without any accompanying safeguards, scope limits, or user-facing warning. This increases the chance of covert browsing, collection of third-party data, or actions on external sites without clear consent or compliance checks.

Content

No source excerpt is available for this finding.