Ai Text To Video Generator

• Outcome: Review
• Latest audit: May 29, 2026 at 9:29 PM UTC
• Version:
  1.2.1

Overview

The skill is a coherent cloud video generator, but it places a live NemoVideo access token inside user-facing links, which makes accidental credential sharing too likely.

Install only if you trust NemoVideo with your prompts and uploaded media, and do not share generated workspace or task links because they contain an active access token. Prefer a version that uses normal login or short-lived one-time claim links without embedding bearer tokens in the URL.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• 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 (5)

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The skill constructs a browser link that embeds the bearer token directly in the URL query string. URLs are commonly exposed via browser history, logs, referrer headers, screenshots, and link sharing, so anyone obtaining that URL may gain access to the user's NemoVideo session and task context.

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The final delivery instructions again require sharing a task-detail link containing the access token, repeating the same credential exposure pattern at the point most likely to be copied or forwarded. Because this is part of normal output delivery, it materially increases the chance of accidental token disclosure and session hijacking.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill instructs automatic token acquisition and local persistence setup on first interaction without clear prior consent or warning. Automatically creating identifiers, requesting remote tokens, and writing to disk on initial contact expands the trust boundary and can surprise users into network and persistence actions they did not knowingly approve.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: Embedding the bearer token in a browser link without a clear warning exposes session credentials through routine URL handling channels. Since the token authenticates API actions, disclosure can let third parties access task data or act within the user's account context.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: Requiring the agent to share a tokenized task-detail link during final delivery operationalizes credential leakage as part of the normal workflow. This makes accidental exposure highly likely because users often save, forward, or paste delivery links outside the trusted context.

VirusTotal

65/65 vendors flagged this skill as clean.