ClawHub
Back to skill

Security audit

sciverse academic retrieval

Security checks across malware telemetry and agentic risk

Overview

This skill is a purpose-aligned Sciverse academic retrieval connector that uses an API token and network calls, with no evidence of hidden persistence, local data harvesting, or destructive behavior.

Install only if you are comfortable sending research queries, paper identifiers, and retrieval requests to Sciverse under your API token. Avoid submitting confidential, regulated, or proprietary research prompts unless your organization approves that provider and data flow.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Missing User Warnings

Low
Confidence
89% confidence
Finding
The README instructs users to configure an API token for a third-party Sciverse service but does not clearly disclose that user queries, prompts, and retrieved document requests will be transmitted off-box to an external provider. This can create an informed-consent and data-handling risk, especially if agents are later used on sensitive or proprietary research queries.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly requires a Sciverse API token and is designed to send search queries, semantic retrieval requests, and content reads to a third-party service, but the documentation does not clearly warn that user prompts, document identifiers, and retrieved content will leave the local environment. This creates a data disclosure and compliance risk, especially if an agent forwards sensitive research topics, proprietary prompts, or regulated data under the assumption that tool use is local or self-contained.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.env_credential_access

Environment variable access combined with network send.

Critical
Code
suspicious.env_credential_access
Location
scripts/_common.mjs:12