Comet Ml

Security checks across malware telemetry and agentic risk

Overview

This looks like a legitimate Comet ML connector, but it gives an agent broad authenticated API access that can change or delete Comet ML data without clear guardrails.

Install only if you are comfortable letting an agent use your connected Comet ML account through Membrane. Use the least-privileged account or workspace available, require explicit approval before any write, delete, bulk, or automation request, and revoke the Membrane connection when you are done.

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 (1)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill explicitly documents a generic proxy request mechanism supporting POST, PUT, PATCH, and DELETE without pairing it with guardrails like confirmation for write operations, least-privilege guidance, or warnings about destructive effects. In an agent setting, this increases the risk that a loosely prompted model performs unintended state-changing operations against the user's Comet ML environment.

VirusTotal

43/43 vendors flagged this skill as clean.

View on VirusTotal