Bibtex to Literature Review

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent document-generation tool that reads user-provided reference files and writes requested DOCX/JSON outputs without evidence of hidden network access, credential use, persistence, or destructive behavior.

Install only if you are comfortable letting the skill read the reference files you provide and write DOCX/JSON outputs to paths you specify. Prefer explicit output directories, review generated bibliographies for accuracy, and avoid enabling external metadata lookup unless you intentionally want web/library enrichment.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill instructs the agent to use shell commands and scripts with file read/write behavior, but no explicit permissions are declared. That mismatch can lead to overbroad execution in environments that rely on declared permissions for policy enforcement or user awareness, especially since the workflow includes creating output files and temporary directories. In this context the behavior appears aligned with the skill’s purpose, so it is not overtly malicious, but it is still a real security and governance issue.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal