Security audit

视频对口型 Video Retalk

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed dLazy VideoRetalk wrapper, with some documentation mistakes but no artifact-backed malicious or review-level behavior.

Before installing, review the pinned @dlazy/cli package and be aware that using this skill sends the media files you choose, plus parameters, to dLazy's cloud service. Prefer DLAZY_API_KEY for per-run credentials if you do not want a saved local config key. Expect possible usage confusion until the stale --prompt and image output examples are corrected.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (7)

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 10)May include surrounding context.
md
[English](./SKILL.md) · [中文](./SKILL-cn.md)

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill is described throughout the manifest as a video lip-sync tool that requires a talking-person video and an audio track, optionally with a reference face image. However, the documented JSON output shows an image/png result, and the command example invokes dlazy videoretalk --prompt ..., which conflicts with the documented required video/audio-style inputs and the skill's stated behavior.
Content: No source excerpt is available for this finding.

Whitespace Padding

Medium

Category: Prompt Injection
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.
Content: Scanner excerpt · SKILL-cn.md (reported line 125)May include surrounding context.
md
## 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` |

Whitespace Padding

Medium

Category: Prompt Injection
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.
Content: Scanner excerpt · SKILL.md (reported line 125)May include surrounding context.
md
## 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` |

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The options block for dlazy videoretalk lists --video_url, --audio_url, and optional --ref_image_url, but the error table says the missing required parameter is --prompt. That is an active contradiction in the file's own documentation and suggests the documentation may have been copied from a different skill.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The manifest and earlier documentation consistently describe this skill as generating a lip-synced video from input video and audio. However, the documented JSON output example shows tool: "videoretalk" returning an output of type: "image" with mimeType: "image/png", which actively conflicts with the stated behavior and could mislead agents about the actual artifact produced.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The options list for dlazy videoretalk documents only video/audio/reference-image related parameters, but the example invocation uses --prompt, and the error table says --prompt is a required option. This is not merely incomplete documentation; it conflicts with the declared lip-sync interface and suggests copy-pasted behavior from a different tool.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.