ClawHub
Back to skill

Security audit

AI Copyright Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent IP document-generation skill, but it can process and retain sensitive project materials in local outputs.

Install only if you intend to generate Chinese IP registration or patent drafting materials. Do not point it at confidential code, unpublished inventions, personal data, credentials, or client materials unless you are authorized and comfortable with local DOCX/PPTX, Markdown, self-check, and temp-script artifacts being retained.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill advertises transforming user-supplied code, papers, and design docs into patent and copyright materials, but it does not clearly warn users that uploaded materials will be analyzed and repurposed into generated artifacts. In an IP-focused workflow, those inputs are likely to contain confidential source code, unpublished inventions, or personal data, so lack of notice raises privacy and data-handling risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list includes generic terms such as 'patent', 'claims', and 'specification' that are common in ordinary conversation and document work. This can cause unintended invocation of the skill on unrelated content, increasing the chance that sensitive code, papers, or internal design documents are processed without deliberate user intent.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
92% confidence
Finding

The skill states that Word output is auto-executed after Phase C unless the user requests Markdown only, meaning file-generation actions occur by default without an explicit final consent step. In this context, autonomous output generation can write sensitive transformed content to disk and create additional artifacts beyond what the user may expect, especially when working with proprietary code or unpublished inventions.

Content

Scanner excerpt · SKILL.md (reported line 247)May include surrounding context.

md
## Phase F: Word Output

Auto-executed after Phase C (unless user requests md only). Uses docx-js workflow: `.temp/{case}-docx.js`.

| Path | Output |
|------|--------|

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill specifies automatic creation of .docx/.pptx files, Markdown intermediates, temp scripts, and self-check reports, but does not warn that sensitive source material may be written to local storage and retained. Because the content involves patents, copyright filings, code excerpts, and desensitization steps, these artifacts may preserve valuable confidential information and create a durable leakage surface.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

Line L48 contains the instruction "AI生成", which indicates the guide is intended to be generated in Chinese or imposes a Chinese-language constraint. The file does not provide any opt-in, language selection, or justification for restricting output language, which can violate language/locale policy requirements.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description states the skill generates Chinese patent applications and CNIPA-format outputs, but it does not present this as a user-selectable locale or justify the restriction in a policy-aware way. This can violate language/locale flexibility expectations when the skill is invoked by users who may expect other languages or jurisdictions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

Line L50 contains the Chinese phrase "AI生成" as a standalone directive/cue. Because the guide does not offer a language choice or explain a required locale, this can be read as forcing a specific language/locale behavior, which falls under the language-policy check.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.