ClawHub
Back to skill

Security audit

ArXiv Summarizer Orchestrator

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent arXiv research-reporting orchestrator that discloses its downloads, local report files, bounded parallel processing, and optional scheduling.

Before installing, make sure the three dependent sub-skills are trusted, pick a bounded output directory, and configure cleanup/retention if running on a schedule because source/PDF downloads and generated markdown reports can accumulate over time.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill orchestrates repeated downloads of paper source/PDF artifacts, writes multiple markdown and metadata files, and is explicitly designed for cron-style scheduled execution, but the user-facing description does not clearly warn about persistent disk writes and recurring network/file activity. This can lead operators to enable or schedule the skill without understanding storage growth, bandwidth use, or the fact that untrusted research artifacts will be fetched and stored locally.

Static analysis

No suspicious patterns detected.