Patent Fee Monitor

• Outcome: Pass
• Latest audit: Jul 20, 2026 · 4:37 AM UTC
• Version: 1.0.0

Overview

This is a coherent patent-fee and IP-asset tracker, with expected local storage and patent-database lookups that users should treat as potentially sensitive.

Install only if you are comfortable storing IP asset records in local files inside the skill directory and sending patent identifiers to public patent lookup services. Avoid using it for highly confidential diligence or pre-publication strategy unless you can control network lookups and protect exported CSV/ICS files.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (8)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill describes behaviors that require network access and reading/writing local files (CSV import/export, ICS generation, external patent lookups) but does not declare corresponding permissions. This creates a capability/expectation mismatch that can bypass user understanding and platform review controls, especially when handling sensitive IP asset data.

Intent-Code Divergence

Medium

• Confidence: 89% confidence
• Finding: The module claims all listed data sources are free and require no API key, but the USPTO path conditionally depends on COZE_USPTO_API_KEY. This misleading disclosure can cause operators or users to misunderstand what external access and credential configuration the skill actually performs.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: Several trigger phrases are broad, such as managing assets or viewing lists, which could cause the skill to activate in contexts where the user did not intend patent-related processing. Unintended invocation is risky here because the skill may process sensitive intellectual property identifiers or initiate external lookups.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill supports exporting IP asset data to CSV and calendar files, but the documentation does not warn that those files may contain sensitive business information such as patent numbers, titles, deadlines, and portfolio structure. Exporting to files or calendar systems can expand data exposure to synced devices, shared folders, backups, and third-party calendar providers.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The workflow states that legal status queries will use external sources such as USPTO, EPO, WIPO, and Google Patents, but it does not warn users that patent numbers and related metadata may be transmitted to third parties. For confidential or pre-publication-sensitive IP management workflows, this can create privacy, compliance, or business-intelligence exposure.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: User-supplied patent numbers are sent to third-party services such as Google Patents without an explicit user warning, consent flow, or clear disclosure. Even if patent numbers are often public, user queries and associated timing can still reveal sensitive business interest, portfolio activity, or internal due-diligence behavior.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill persists asset records to local files and also supports export/delete operations without clearly warning users that data will be written to disk. Because the data includes IP asset identifiers, ownership, dates, and notes, silent persistence can create confidentiality and integrity risks on shared or unmanaged environments.

External Transmission

Medium

• Category: Data Exfiltration

Content

"limit": 1,
            "offset": 0
        }
        resp = requests.post(url, headers=headers, json=payload, timeout=15)
        
        if resp.status_code != 200:
            return None

• Confidence: 93% confidence
• Finding: requests.post(url, headers=headers, json=

VirusTotal

65/65 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.