Paper Impact Analyzer

• Outcome: Pass
• Latest audit: May 29, 2026, 8:29 PM
• Version:
  1.1.0

Overview

This is a coherent paper-analysis skill that contacts public research APIs, but its current script weakens HTTPS verification and should not be used for confidential research queries.

Reasonable to install for public academic paper checks. Do not use it for confidential, embargoed, or sensitive research interests unless you are comfortable sending the arXiv ID and related metadata to arXiv, GitHub, OpenAlex, and Semantic Scholar. Treat the results as advisory because the current script disables HTTPS verification, making network tampering possible.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 94% confidence
• Finding: The skill declares only Bash/Read/Write tools, but its documented behavior and required implementation clearly depend on outbound network access to arXiv, GitHub, OpenAlex, and Semantic Scholar. This creates a permissions mismatch: a reviewer or execution environment may believe the skill is local-only while it actually performs external requests, which can bypass expected trust boundaries, leak user queries, and violate least-privilege assumptions.

Intent-Code Divergence

Medium

• Confidence: 99% confidence
• Finding: The code globally disables TLS certificate validation and hostname verification, then reuses that insecure SSL context for all HTTPS requests. This allows man-in-the-middle interception or tampering of responses from GitHub, OpenAlex, and Semantic Scholar, so the tool can be fed falsified metadata or malicious content while appearing to use HTTPS.

Intent-Code Divergence

Medium

• Confidence: 99% confidence
• Finding: The script globally disables TLS certificate validation and hostname checking, so every HTTPS request to GitHub, OpenAlex, and Semantic Scholar becomes vulnerable to man-in-the-middle interception or response tampering. In this skill, untrusted network responses directly influence analysis output, so an attacker on the network path could falsify repo metrics, citation counts, or metadata and potentially harvest queried data.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill sends user-supplied arXiv IDs, derived paper metadata, and first-author names to multiple third-party services without explicit notice or consent at runtime. While this is part of the feature design, it creates a privacy and data-governance risk because users may not expect their research interests or queried authors to be disclosed to external APIs.

VirusTotal

65/65 vendors flagged this skill as clean.