Agently Mail

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be an email helper with no malware telemetry or artifact-backed malicious behavior, but its trigger wording is broader than ideal for sensitive email actions.

Install only if you expect the agent to operate on your email account. Before use, confirm what mailbox permissions the connected tool has, and require explicit confirmation before reading broad mailbox content or sending/replying/forwarding messages.

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: 94% confidence
Finding: The skill description says to use this skill for 'any email-related operation,' which is overly broad and can overlap with ordinary conversational requests about email. In an agent environment, that increases the chance of unintended invocation, leading to premature access to mailbox data or initiation of email-sending flows when the user only wanted discussion or drafting help.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal