博客转视频 Blog to Video

• Outcome: Review
• Latest audit: Sep 4, 2026 · 1:24 AM UTC
• Version: 1.0.11

Overview

The skill is not clearly harmful, but its name understates a broader document-to-video workflow that uploads prompts and selected files to dLazy and stores a local API key.

Review this as a general document-to-video connector, not only a blog-to-video skill. Install it only if you are comfortable using the dLazy npm CLI, storing a dLazy API key locally, and sending prompts plus any explicitly attached files to dLazy services.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Ae1

High

• Category: analysis-evasion

Content

[English](./SKILL.md) · [中文](./SKILL-cn.md)

• Confidence: 100% confidence
• Finding: Referenced artifact was not completely inspected

Intent-Code Divergence

Medium

• Confidence: 94% confidence
• Finding: The top-level title says 'Blog to Video,' but the body immediately broadens the behavior to generic document-to-video processing and references `file-to-video`. This misleading capability description undermines informed consent and can trick routing logic or users into sending unrelated files to an external service under a narrower-looking label.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The trigger list contains broad terms such as document-to-video, PPT, PDF explanation, courseware video, and report interpretation, which substantially exceed the named skill scope. Overbroad triggers can cause accidental invocation for common requests, leading users to send files or content to a third-party API without realizing they selected a broader external-processing workflow.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The documentation states that messages are sent to `api.dlazy.com` and local files are uploaded to `files.dlazy.com`, but it does not place a prominent, up-front privacy warning before use. This is risky because users may provide sensitive blog drafts, internal documents, or media without clear notice that their content leaves the local environment and is retained in a SaaS project context.

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The skill metadata advertises a narrow blog/article-to-video capability, but the instructions invoke a broader `file-to-video` template that can process many other document types. This scope mismatch can cause the agent to apply the skill to inputs the user did not intend, increasing the chance of unnecessary file handling and external transmission to the vendor service.

Whitespace Padding

Medium

• Category: Prompt Injection

Content

## 错误处理

| Code | 错误类型                         | 示例信息                                                                                                                 |
| ---- | -------------------------------- | ------------------------------------------------------------------------------------------------------------------------ |
| 401  | 未授权 (API Key缺失或无效)       | `ok: false, code: "unauthorized", message: "API key is missing or invalid"`                                                              |
| 501  | 缺少必填参数                     | `error: required option '--prompt <prompt>' not specified`                                                               |

• Confidence: 70% confidence
• Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The manifest description at L004 scopes the skill to converting a blog post or article into a video. However, the skill body says it handles PPT, Word, Excel, PDF, and generic documents, and the usage section shows the actual backend template is `file-to-video`, indicating a broader document-to-video capability than the manifest claims.

Whitespace Padding

Medium

• Category: Prompt Injection

Content

## Error Handling

| Code | Error Type                         | Example Message                                                                                                          |
| ---- | ---------------------------------- | ------------------------------------------------------------------------------------------------------------------------ |
| 401  | Unauthorized (No API Key)          | `ok: false, code: "unauthorized", message: "API key is missing or invalid"`                                                              |
| 501  | Missing required parameter         | `error: required option '--prompt <prompt>' not specified`                                                              |

• Confidence: 70% confidence
• Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.