ClawHub
Back to skill

Security audit

paper-digest

Security checks across malware telemetry and agentic risk

Overview

This skill behaves as a SoMark-based paper parser and research-card helper, with no hidden or unrelated behavior found, but it does send selected documents to SoMark.

Install only if you are comfortable sending the chosen paper or document to SoMark using your SOMARK_API_KEY. Do not use it for confidential, unpublished, regulated, or proprietary documents unless SoMark's terms and data handling are acceptable to you. Confirm the base URL is an official SoMark endpoint before running.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill advertises use of environment variables, local file reads/writes, and network access through shell commands, but does not declare corresponding permissions. This creates a transparency and consent problem: users and enforcement layers may underestimate the skill’s actual access to files and external services, increasing the chance of unintended data exposure when processing local documents and sending them to SoMark.

Tp4

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding
The documented behavior materially overstates what the skill does and understates what it can process. A description/behavior mismatch is dangerous because users may provide sensitive files believing they will receive local AI analysis, while the skill actually acts as a remote parsing client and supports additional file types such as .doc/.docx that broaden the ingestion surface.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The tool unconditionally uploads the user-supplied paper bytes and API key to a remote SoMark endpoint, but provides no explicit warning, confirmation, or privacy notice at the point of use. In a skill context, this can expose unpublished papers, proprietary documents, or sensitive research data to a third-party service without sufficiently informed user consent.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.