Chift

Security checks across malware telemetry and agentic risk

Overview

This looks like a legitimate Chift/Membrane integration, but it needs review because it gives an agent broad authenticated power to run actions and raw API requests against connected business systems.

Install only if you trust Membrane/Chift with the connected account. Use a least-privileged or test account where possible, prefer listed Membrane actions over raw proxy requests, and require the agent to ask before any create, update, delete, workflow change, or direct proxy request.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The skill description is overly broad ('Manage data, records, and automate workflows'), which can cause the agent to invoke this skill for generic business or data tasks beyond a clearly bounded Chift-specific use case. That increases the chance of unintended networked actions against external systems, especially because the skill supports creating connections, running actions, and proxying arbitrary API requests.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal