ClawHub
Back to skill

Security audit

patent-research-analyst

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed patent-search and analysis workflow that uses PatSnap/Zhihuiya patent tools after user confirmation, with no executable code or persistence in the artifact.

Before installing, users should understand that this skill is for patent research and FTO-style analysis using PatSnap/Zhihuiya MCP tools. Treat its legal-risk and infringement outputs as research support, not a substitute for legal advice, and only provide technical details or competitor names that are appropriate to share with the configured patent data service.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list includes broad natural-language phrases such as ‘帮我做专利检索’, which are common request phrasings and may activate the skill in contexts where the user did not intend the full patent-analysis workflow. In an agent environment with tool access, unintended activation can lead to unnecessary data access, confusing workflow hijacking, and generation of legal-risk conclusions without clear user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation conditions are broad and lack exclusion criteria, so the skill may start on loosely related requests like general technology analysis, competitive intelligence, or research planning that are not necessarily patent-search tasks. Because this skill performs multi-step collection, retrieval, and legal-risk analysis, ambiguous activation increases the chance of over-collection, misrouting the conversation, and producing sensitive or high-stakes outputs without sufficiently explicit user authorization.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.