Security audit

全能生成 Generate

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently supports media generation and clearly discloses its cloud uploads, API key storage, and installation requirements.

Use it for media requests you intend to send to dLazy, consider generation charges and the service's privacy terms, and protect the revocable API key saved locally. The external CLI executable was not fully audited in this review.

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
Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

The standalone trigger “生成” (“generate”) is too broad for a media-generation skill.

Medium

Category: Trigger Abuse
Confidence: 97% confidence
Finding: “生成” commonly appears in requests to generate text, code, reports, and other non-media outputs. Although the description identifies image, video, and audio generation as the skill's purpose, the trigger list does not explicitly restrict this standalone keyword to those contexts or provide exclusions.
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 189)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` |

The standalone trigger "generate" is overly broad and can match requests unrelated to media generation.

Medium

Category: Trigger Abuse
Confidence: 97% confidence
Finding: The trigger list includes "generate" without requiring an image, video, or audio context. This common verb also appears in requests to generate code, reports, or other non-media content, and no explicit exclusion conditions constrain the trigger.
Content: No source excerpt is available for this finding.

The voice-search instructions require an English search descriptor without user opt-in or a documented English-only API constraint.

Low

Category: Output Handling
Confidence: 86% confidence
Finding: The instruction “只给一个英文描述词检索” explicitly requires the agent to produce an English search term. The explanation justifies using a single term rather than multiple words, but does not explain why English is required or offer a language choice.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.