ClawHub
Back to skill

Security audit

Research Paper Reading

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only research paper reading aid with disclosed Zotero lookup examples and no evidence of hidden execution, persistence, or data exfiltration.

Installers should expect this skill to guide paper-reading output in a largely Chinese template style. If using the Zotero examples, only run them when you intend the agent to read local Zotero item or collection metadata from the localhost API.

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)

Natural-Language Policy Violations

Medium
Confidence
97% confidence
Finding
The skill hard-codes Chinese section headers and reading/output structure without any indication that language should follow the user's preference. In an agent setting, this can override user intent, degrade usability, and create instruction-priority conflicts where the model produces responses in an unintended language.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding
Suspicious Unicode normalization or mixed-script content

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
This markdown template is predominantly written in Chinese, but line L193 introduces an English heading ('Relevance to My Research') with no user choice or explanation. That creates a mild language/locale policy inconsistency because the skill content implicitly fixes presentation language rather than clearly offering or documenting the language mode.