Back to skill

Security audit

数字人动作视频 即梦 Dream Actor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed dLazy cloud media-generation wrapper, with ordinary SaaS and npm supply-chain risks but no artifact-backed malicious behavior.

Install this only if you are comfortable using dLazy's cloud service for prompts and selected media files. Prefer the npx/on-demand path or an isolated local install over a global install, review the @dlazy/cli source/package before use, avoid running it as administrator, and keep the dLazy API key revocable and rotated if needed.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:4
Finding
Execution of an External npm Package Without Artifact Integrity Verification<![CDATA[ ## Vulnerability Details **File Location**: `SKILL.md:4,49-55`; `SKILL-cn.md:4,49-55` **Vulnerability Type**: Third-party supply-chain exposure **Risk Level**: Medium ### Vulnerable Code `SKILL.md`: ```yaml metadata: {"clawdbot":{"emoji":"🤖","requires":{"bins":["npm","npx"]},"install":"npm install -g @dlazy/cli@1.2.3","installAlternative":"npx @dlazy/cli@1.2.3","homepage":"https://github.com/dlazy-ai/cli","source":"https://github.com/dlazy-ai/cli","author":"dlazyai","license":"see-repo","npm":"https://www.npmjs.com/package/@dlazy/cli","configLocation":"~/.dlazy/config.json","apiEndpoints":["api.dlazy.com","files.dlazy.com"]},"openclaw":{"systemPrompt":"When invoking this skill, use dlazy jimeng-dream-actor -h for help."}} ``` ```bash npx @dlazy/cli@1.2.3 <command> ``` ```bash npm install -g @dlazy/cli@1.2.3 ``` `SKILL-cn.md`: ```yaml metadata: {"clawdbot":{"emoji":"🤖","requires":{"bins":["npm","npx"]},"install":"npm install -g @dlazy/cli@1.2.3","installAlternative":"npx @dlazy/cli@1.2.3","homepage":"https://github.com/dlazy-ai/cli","source":"https://github.com/dlazy-ai/cli","author":"dlazyai","license":"see-repo","npm":"https://www.npmjs.com/package/@dlazy/cli","configLocation":"~/.dlazy/config.json","apiEndpoints":["api.dlazy.com","files.dlazy.com"]},"openclaw":{"systemPrompt":"当调用此技能时,可以使用 dlazy jimeng-dream-actor -h 查看帮助信息。"}} ``` ```bash npx @dlazy/cli@1.2.3 <command> ``` ```bash npm install -g @dlazy/cli@1.2.3 ``` ### Technical Analysis The Skill delegates its core behavior to the external npm package `@dlazy/cli@1.2.3`. Pinning the package version limits unintended version drift, but the project does not vendor the executable implementation or provide a cryptographic checksum, signature, lockfile, or other mechanism for independently verifying the retrieved package artifact. Consequently, the supplied project cannot establish that the code executed during installation and invocation is identical to code previously reviewed. npm ins ...[truncated 2168 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Vendor and review the exact CLI implementation and its required dependencies as part of the audited Skill artifact where licensing and maintenance constraints permit. 2. Publish cryptographic checksums or signed provenance for approved package artifacts and verify them before execution. 3. Use an npm lockfile with integrity fields for a controlled installation workflow, including pinned transitive dependencies. 4. Prefer a project-local, isolated installation over `npm install -g` so the binary does not persist globally or affect unrelated workflows. 5. Execute the CLI in a sandbox or container with access limited to: - The specific input media selected by the user. - A dedicated output directory. - The minimum required configuration file. - Only the documented dLazy network endpoints. 6. Avoid running npm or the CLI with administrator or root privileges. 7. Disable npm lifecycle scripts where compatible with the verified package behavior, or separately audit every required lifecycle script before allowing execution. 8. Protect API credentials using a platform credential store where possible, and provide narrowly scoped, revocable keys. 9. Establish dependency monitoring and a documented process for reviewing package ownership changes, newly disclosed vulnerabilities, and transitive dependency updates. 10. Apply equivalent changes to both `SKILL.md` and `SKILL-cn.md` so their installation guidance remains consistent. ]]>
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (7)

Ae1

High
Category
analysis-evasion
Content
[English](./SKILL.md) · [中文](./SKILL-cn.md)
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger keywords include broad phrases like '图片生成视频' and '让图片动起来', which can cause the agent to invoke this skill unintentionally in ordinary conversation. Because the skill sends prompts and local file paths to a third-party SaaS and may upload referenced local media, accidental invocation can lead to unintended data transfer, unwanted API usage, and confusing autonomous behavior.

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.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The trigger keywords include broad, generic phrases like "image to video" and "make image move," which are likely to match benign user requests that do not specifically intend to invoke this third-party cloud skill. That raises the risk of unintended invocation, causing user data such as prompts and local media paths/files to be sent to external dLazy services without sufficiently explicit user intent.

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.

Description-Behavior Mismatch

Low
Confidence
95% confidence
Finding
The manifest and description repeatedly state that this skill converts a static character image into an action video. However, the documented output example shows `tool: "jimeng-dream-actor"` returning an output with `type: "image"`, `mimeType: "image/png"`, and a `.png` URL, which does not match the stated video-generation purpose.

Natural-Language Policy Violations

Low
Confidence
80% confidence
Finding
SQP-3 适用于所有文件类型。这里的指令使用“您必须明确告知用户”等强制性措辞规定输出行为,但未说明应根据用户语言偏好或当前对话语言进行调整,可能构成语言/locale 强制要求。

Static analysis

No suspicious patterns detected.