Back to skill

Security audit

胡田 OPC导师 AI资产知识产权保护

Security checks for vulnerabilities and agentic risk

Overview

This skill is not malware, but it contains high-stakes IP filing guidance that could lead users to submit misleading legal materials.

Install only if you will independently verify all legal filing requirements with the relevant authority or counsel. Do not sign any no-AI or originality certification unless it is true for your materials, and do not pad code or documents just to satisfy submission thresholds. Store IP ledgers, IDs, source code, certificate numbers, and knowledge-base identifiers in access-controlled locations.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (9)

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manual instructs applicants to sign a commitment that submitted materials were not AI-generated, despite the skill being positioned as protecting AI-related成果. More importantly, this directs users toward making a potentially false legal certification, which can expose them to fraud, rejection, or liability if AI assistance was actually used. In the context of an IP-protection skill, this is more dangerous because users are likely to rely on the document for compliance-sensitive filings.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The troubleshooting section recommends inflating code length by adding comments, blank lines, or auxiliary code to satisfy filing thresholds. This encourages users to manipulate submission artifacts rather than provide an authentic representation of the software, creating risk of deceptive filing, rejection, or legal consequences. In a legal/IP filing guide, advice to pad evidence is especially dangerous because it targets regulated submissions where truthfulness is material.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest description is written entirely in Chinese, and the document consistently specifies behavior and output only in Chinese without indicating that other languages are supported. Under the policy, a skill that imposes a specific language without user choice is a natural-language policy violation unless the locale restriction is clearly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The section instructs the skill to output a fixed Chinese template when triggered, and it does not provide any opt-in or alternative for other languages. This is a direct language policy concern because it constrains responses to a specific language regardless of user preference.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Line L069 contains a template statement requiring the applicant to promise that the submitted materials "未使用人工智能(以下简称AI)自动生成". This is a natural-language policy constraint about AI usage presented as a blanket requirement, without explaining a justified compliance context or offering any user opt-in or alternative path.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The document explicitly recommends exporting the IP registry to Excel/CSV, using cloud document synchronization, and printing key information, but it does not pair those actions with any confidentiality controls. In the context of a ledger tracking trade secrets, prompt logic, certificate numbers, storage paths, and knowledge-base identifiers, these recommendations materially increase the risk of unauthorized disclosure or wider internal spread of sensitive IP data.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
76% confidence
Finding

Line L099 states '承诺未使用AI生成', which is a natural-language policy assertion about disallowing AI-generated material. In this file, that restriction is presented as a requirement without any opt-in, alternative path, or contextual justification, which can constitute a policy-style constraint embedded in the skill content.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The README presents all user-facing guidance in Chinese and does not mention that the skill is Chinese-only, region-specific, or that users may choose another language. This can violate language/locale policy expectations where user opt-in or explicit locale justification is required.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

This file is entirely written in Chinese and frames the guidance as universally applicable, but it does not explicitly state that the content is intended only for Chinese software copyright registration workflows or offer any language choice. Under the policy, forcing a specific language/locale without user opt-in can be a natural-language policy issue unless the locale constraint is clearly documented and justified.

Content

No source excerpt is available for this finding.