Back to skill

Security audit

Iris Lite — Quick Inbox Scan

Security checks across malware telemetry and agentic risk

Overview

This Gmail inbox triage skill does what it says, but users should understand it requires a Gmail app password and reads recent email data.

Install only if you are comfortable granting IMAP access to the Gmail inbox configured in the environment. Use a dedicated Gmail app password, revoke it when finished if desired, and expect recent email headers plus short snippets to be processed and priority information to be printed locally.

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

Medium
Confidence
90% confidence
Finding
The skill explicitly advertises scanning a user's Gmail inbox and requires a Gmail app password, but it does not present a clear privacy warning, data-handling disclosure, or limits on what inbox data is accessed and displayed. Because this processes sensitive mailbox contents and credentials, users may not fully understand the exposure before running it.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The manifest requests a highly sensitive credential, GMAIL_APP_PASSWORD, without an explicit warning about safe handling, storage, and scope of that secret. Requesting mailbox credentials in a skill materially increases risk because compromise of the environment or misuse of the skill can expose the entire mailbox.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:81