ClawHub
Back to skill

Security audit

蒋道理｜CNIPA商标驳回复审证据推理引擎（SJ-IRAC）

Security checks for vulnerabilities and agentic risk

Overview

This is a CNIPA trademark refusal-review drafting skill with legal-risk caveats to consider, but it has no executable code, hidden behavior, persistence, credential use, or automated filing mechanism.

Install only if you need PRC/CNIPA trademark refusal-review drafting. Do not treat Document Mode output as a final filing without review by a qualified trademark professional, especially for deadlines, standing, evidence admissibility, and jurisdiction-specific strategy.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill explicitly markets output as '可直接提交' and 'examiner-readable' legal review materials, which can encourage users to file AI-generated arguments without qualified human review. In a legal-adjudicative context, missing or weak warnings materially increase the risk of inaccurate claims, procedural mistakes, waived arguments, or jurisdiction-specific noncompliance being submitted to a government authority.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

Line L020 states the skill should 'Always apply PRC Trademark Law (2019 Amendment)' unless a temporal rule requires otherwise. This is a locale/jurisdiction constraint expressed as a blanket rule, but the document does not explicitly frame the skill as PRC-only or offer a user choice of jurisdiction, which can create a language/locale policy issue.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The risk dimensions use Chinese parenthetical descriptions embedded in an otherwise English document, which imposes a language requirement on readers without any opt-in or stated region-specific justification. This can violate language/locale policy because users are not given a choice of language and the constraint is not documented.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.