Security audit

视频抠像 Video Segmentation

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed wrapper for a hosted video segmentation CLI, with documentation issues but no artifact-backed evidence of hidden, destructive, or deceptive behavior.

Before installing, treat this as a third-party cloud CLI: review the pinned npm package/source if you rely on it, expect user-provided media files to be uploaded to dLazy endpoints, and store the API key only if that persistence is acceptable. The publisher should fix the `--prompt` and image/video documentation mismatches to avoid failed or misprocessed calls.

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 (9)

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: 89% confidence
Finding: Manifest 和正文多次说明该技能返回“与原视频同时长的 mask 视频”。然而输出示例中的 outputs[0] 被标记为 type: "image"、mimeType: "image/png"，这与“视频分割返回 mask 视频”的描述形成直接冲突，可能导致使用方错误处理结果类型。
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 121)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` |

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The agent instruction text mandates that responses to certain error conditions '必须明确告知用户' using Chinese-language guidance, but this Chinese skill file does not state that the user has opted into Chinese-only responses. That creates a locale/language policy concern because the behavior is prescribed rather than offered as a choice.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The manifest and main description repeatedly say this skill performs video human segmentation and returns a same-length mask video. However, the documented JSON output shows an output object with type: "image", url: "...result.png", and mimeType: "image/png", which conflicts with the stated video output behavior in a security-relevant way because it misrepresents what artifact is produced.
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.md (reported line 121)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: 95% confidence
Finding: 该技能整体描述为“视频人像分割”，帮助信息中也只展示了 --video 等与视频处理相关的参数。但命令示例写成 dlazy videoseg --prompt '提示词内容'，与前述接口用途和参数定义直接不一致，容易误导调用方按文本生成类用法使用该技能。
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 97% confidence
Finding: 帮助信息仅定义了 --video、--dry-run、--no-wait、--timeout、--save 等参数，没有显示 --prompt。但错误处理表中把“缺少必填参数”的示例写成 required option '--prompt <prompt>' not specified，这与该技能宣称和展示的实际命令接口相冲突。
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The usage section lists only --video, --dry-run, --no-wait, --timeout, and --save, but the example invokes dlazy videoseg --prompt 'prompt content' and the error table says --prompt is required. This is an active contradiction within the skill documentation about what inputs the command accepts and requires.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.