Security audit

OPC 资源导航

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Chinese-language tool recommendation guide with no executable code, hidden persistence, or instructed data exfiltration, though users should approve any browsing or file output.

Before installing, expect a Chinese-language recommendation assistant that may read its bundled tools database and, when available, browse official product sites to verify prices. Ask it to confirm before browsing or saving a report, and treat the bundled automation image as an explanatory example rather than instructions to use cookies, run scripts, or publish content automatically.

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
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 (5)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger list includes broad everyday phrases such as “推荐工具”, “用什么工具”, and “我该学什么”, which can cause the skill to activate in many unrelated contexts. This increases the chance of unintended file reads, browsing behavior, or recommendation generation when the user did not explicitly ask for this specific skill.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill says the result may be written to a file, but it does not require explicit user consent, identify the destination, or warn that persistent data will be created or modified. That creates a risk of unintended local data modification, overwriting existing content, or storing sensitive business information without the user's awareness.

Natural-Language Policy Violations

Medium

Confidence: 87% confidence
Finding: The user-facing metadata is entirely in Chinese ("OPC 工具与资源导航" and the Chinese short description), which indicates a fixed language choice in the skill interface. The file does not offer a language choice or document a region-specific justification, so this is a natural-language locale policy concern.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: This JSON contains user-facing natural-language descriptions, goals, and selection rules entirely in Chinese. Under the policy, forcing a specific language without user opt-in or a documented regional justification is a natural-language policy violation.

Missing User Warnings

Low

Confidence: 80% confidence
Finding: The skill instructs the agent to use research or browsing tools to visit product websites for price and availability checks, but it does not tell the user that external network access may occur or that query context could be exposed to third parties. In a recommendation skill, this is not inherently malicious, but the missing transparency can lead to privacy surprises and unnecessary outbound requests.

Static analysis

No suspicious patterns detected.