Moji Vocab

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:15 PM UTC
• Version:
  1.0.0

Overview

This skill is a coherent Moji vocabulary helper, but it needs a Moji session token and can delete saved vocabulary when the user runs deletion commands.

Install only if you trust the publisher and are comfortable giving the local scripts your Moji session token and device ID. Store those values locally, do not paste them into shared chats or logs, and run deletion workflows with --dry-run first because removed vocabulary may not be easy to recover.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill explicitly instructs users to export sensitive session credentials from browser storage and uses network access to call a third-party API, yet no permissions or trust boundaries are declared. This creates a real security issue because users are encouraged to grant powerful capabilities and provide secrets without transparent disclosure, making accidental credential misuse or overreach more likely.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill tells users to retrieve sessionToken and deviceId directly from browser local storage and place them into environment variables, but provides no warning that these values are effectively account credentials. Exposing and reusing active session material increases the risk of account takeover if logs, shell history, screenshots, shared terminals, or downstream tooling leak the values.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill advertises deletion of saved vocabulary items without a clear warning, confirmation model, or rollback guidance. In this context, deletion affects user study data and could cause irreversible loss if triggered incorrectly, especially since the skill also appears capable of broader deletion behavior than the description implies.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: The delete-by-level action performs real, irreversible deletion immediately when invoked, with no interactive confirmation, preview of affected items, or safety interlock beyond an optional dry-run flag. In a vocabulary-management skill, destructive actions are expected, but the lack of confirmation materially increases the chance of accidental data loss from operator error or mis-specified levels.

VirusTotal

65/65 vendors flagged this skill as clean.