ClawHub
Back to skill

Security audit

Paper Parse Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is a paper-parsing helper, but it asks for broad write and command execution authority and includes underspecified file mutation and callback behavior.

Review this skill before installing. Use it only with papers and directories you are comfortable exposing to an agent, avoid supplying callback URLs unless you trust the destination, and require confirmation before any file modification, deletion, package installation, or command execution.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill is presented as a narrowly scoped paper/PDF parsing tool, but the documentation expands its behavior to generic create/query/modify/delete/export operations. That scope mismatch is dangerous because it can normalize broader agent actions than users expect, increasing the chance of unauthorized file manipulation or data handling under the cover of a benign parsing skill.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill declares exec capability and explicitly classifies itself as MD+EXEC without establishing a clear need for arbitrary command execution in a paper parsing workflow. In agent environments, unnecessary exec access materially raises the risk of command injection, filesystem abuse, environment discovery, and execution of unsafe shell commands derived from user-controlled inputs or document content.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill exposes write and exec capabilities but does not warn users that it may modify files or run commands. This is dangerous because users may invoke what appears to be a simple document parser without understanding that the skill can change local state or execute system-level actions, undermining informed consent and safe operation.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documented callback_url feature enables network transmission of processing results, but there is no warning about privacy, data egress, or trust boundaries. For a paper parsing skill, uploaded documents may contain unpublished research or sensitive metadata, so silent callbacks could leak confidential content to attacker-controlled endpoints.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.