Back to skill

Security audit

PDF 转视频 PDF to Video

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed wrapper around dLazy's CLI/SaaS for turning documents into videos; it uploads selected files and uses a local API key, but no hidden or destructive behavior was found.

Install only if you are comfortable sending prompts and attached documents to dLazy's hosted service and using the @dlazy/cli npm package. Prefer the npx path if you do not want a global binary, review the CLI source/package provenance for your risk tolerance, and confirm the skill is intended before uploading sensitive documents.

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:5
Finding
Unverified Third-Party CLI Is Downloaded and Executed<![CDATA[ ## Vulnerability Details **File Location**: `SKILL.md:5, 65-69`; equivalent translated instructions in `SKILL-cn.md:5, 61-65` **Vulnerability Type**: `T08: Insecure Dependencies` **Risk Level**: Medium ### Vulnerable Code From `SKILL.md:5`: ```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 this skill is called, run 'dlazy chat --skill file-to-video --prompt ...' for a new task, or 'dlazy chat --project <id> --prompt ...' to continue (discover ids via 'dlazy projects list'). Never pass both --skill and --project."}} ``` From `SKILL.md:65-69`: ```markdown You can install on demand without persisting a global binary by running: ```bash npx @dlazy/cli@1.2.3 <command> ``` ``` The translated file contains the same behavior at `SKILL-cn.md:5, 61-65`: ```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,新任务运行 'dlazy chat --skill file-to-video --prompt ...',继续已有项目用 'dlazy chat --project <id> --prompt ...'(用 'dlazy projects list' 查 id);不要同时传 --skill 和 --project。"}} ``` ```bash npx @dlazy/cli@1.2.3 <command> ``` ### Technical Analysis The Skill directs the agent or user to install or execute `@dlazy/cli@1.2.3` from the npm registry ...[truncated 2766 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Bundle the required implementation with the Skill so that its executable code can be reviewed alongside the Skill documentation. 2. If external package retrieval remains necessary, publish and verify a cryptographic integrity value for the exact npm artifact before execution. 3. Provide a lockfile or equivalent manifest that pins the complete transitive dependency graph rather than only the top-level package version. 4. Establish provenance between the npm artifact and a specific reviewed source revision through signed releases, npm provenance attestations, and reproducible builds. 5. Prefer ephemeral, least-privilege execution in a sandbox or container instead of global installation. 6. Disable npm lifecycle scripts where operationally possible, for example by using an installation process that applies `--ignore-scripts` after verifying that the package does not legitimately require them. 7. Restrict filesystem and network access to the minimum required destinations and documents. 8. Keep API credentials scoped, revocable, and short-lived where possible, and ensure that credential files remain readable only by the owning user. 9. Document a package-verification procedure that users or agents must complete before first execution and after every dependency update. 10. Independently audit the exact published package and its complete dependency graph before treating it as trusted. ]]>

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (3)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger list contains broad phrases such as 'document to video' and 'explainer video' that could match ordinary user requests and invoke the skill unintentionally. In this skill's context, unintended invocation is more sensitive because it can cause local files to be uploaded to third-party services and send prompts/content to external APIs, creating privacy, cost, and workflow-integrity risks.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger keywords are broad generic phrases such as 'document to video' and 'explainer video', which can match many common user requests unrelated to this specific tool. In an agent environment, overly broad activation can cause unintended routing to this skill, leading users to upload documents or send content to an external SaaS when they did not intend to use this provider.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The invocation description expands scope from PDF to many document types ('ppt', 'word', 'excel', 'document to video'), making the activation boundary unclear. That ambiguity increases the chance the agent will invoke this skill for broad document-help requests, which is more dangerous here because the skill sends prompts and attached local files to remote endpoints (api.dlazy.com and files.dlazy.com).

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.