ClawHub
Back to skill

Security audit

Ip Quick Screen

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed patent/IP screening helper that queries a public patent API and does not show hidden persistence, credential access, destructive behavior, or unrelated data handling.

Install only if you are comfortable sending searched company names, patent numbers, and related patent-family identifiers to the USPTO PatentsView service. Avoid using it for confidential deal targets or sensitive diligence strategy unless that external lookup is acceptable in your environment.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (11)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill describes use of external patent data sources and implicitly requires network access, but it does not declare any explicit tool scope or permission boundaries. This can lead to over-broad agent execution, making it harder to constrain outbound requests or prevent unintended access patterns if the runtime grants default network or environment capabilities.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger description is broad enough to match many generic requests about risk assessment, analysis, due diligence, or patents, which can cause the skill to activate in contexts the user did not intend. Over-broad activation increases the chance of unnecessary external queries, inadvertent disclosure of user-provided company or patent data to third-party services, and inappropriate reliance on a quick-screening workflow for higher-stakes legal or diligence tasks.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The file’s natural-language descriptions and all printed/report text are fixed in Chinese, which imposes a specific language on users. There is no mechanism to select locale or explicitly document that the skill is intended only for Chinese-speaking users in a constrained regional context.

External Transmission

Medium
Category
Data Exfiltration
Content
# 常量定义
# ============================================================

USPTO_API_URL = "https://api.patentsview.org/patents/query"

# IPC大类 - 技术领域映射（常用）
IPC_SECTION_MAP = {
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
}

    try:
        resp = requests.post(
            USPTO_API_URL,
            json=query,
            timeout=30,
Confidence
88% confidence
Finding
This request transmits user-controlled query content to an external third-party API. Even though the destination is a legitimate patent data service over HTTPS, the transmission can leak sensitive due-diligence targets and research interests outside the local environment.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill sends user-supplied company names or patent numbers to the external USPTO PatentsView API without any explicit disclosure, consent step, or data-handling notice. In a due-diligence context, those queries can reveal confidential investigation targets, deal activity, or proprietary interest, so silent transmission creates a real privacy and business-sensitivity risk.

External Transmission

Medium
Category
Data Exfiltration
Content
}

    try:
        resp = requests.post(
            USPTO_API_URL,
            json=query,
            timeout=30,
Confidence
88% confidence
Finding
This code sends patent-number-based lookups to an external USPTO endpoint. In an IP diligence workflow, even patent-number lists may be confidential because they can disclose the exact assets under review or strategy focus.

External Transmission

Medium
Category
Data Exfiltration
Content
],
                "o": {"per_page": len(batch), "page": 1}
            }
            resp = requests.post(
                USPTO_API_URL,
                json=query,
                timeout=15,
Confidence
84% confidence
Finding
The helper performs additional external requests for family information, increasing the amount of data sent and broadening metadata exposure beyond the user's initial expectation. Chained enrichment calls can amplify confidentiality risk because more identifiers are transmitted automatically.

External Transmission

Medium
Category
Data Exfiltration
Content
"f": ["patent_number", "patent_family_id"],
                "o": {"per_page": 200, "page": 1}
            }
            resp = requests.post(
                USPTO_API_URL,
                json=query,
                timeout=15,
Confidence
84% confidence
Finding
This second-stage family-ID query performs further outbound transmission to enumerate related patents, which can reveal additional analysis scope and internal diligence patterns. The risk is contextual rather than exploitative, but it is still a genuine external data exposure concern.

Natural-Language Policy Violations

Low
Confidence
80% confidence
Finding
All user-facing natural-language instructions and examples are presented only in Chinese, with no indication that the skill can adapt to the user's preferred language. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern unless the locale restriction is explicitly justified.

Natural-Language Policy Violations

Low
Confidence
84% confidence
Finding
The only natural-language content in this file is written in Chinese ('依赖', '无需额外依赖'), which can indicate a fixed language choice without user opt-in. Because no justification or multilingual option is provided in the file, this may conflict with language/locale policy expectations.

Static analysis

No suspicious patterns detected.