Back to skill

Security audit

视频生成 MiniMax H3

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed dLazy video-generation wrapper, with ordinary SaaS data sharing and npm supply-chain risk users should understand before installing.

Install only if you trust the dLazy CLI package and service. Prefer the `npx` invocation when you do not want a global executable, use a revocable dLazy API key, and pass only media files you intend to upload to dLazy.

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
Execution of a Third-Party npm Package Without Locally Auditable Implementation or Integrity Verification<![CDATA[ ## Vulnerability Details **File Location**: `SKILL.md:5, 53-60`; equivalent installation directives in `SKILL-cn.md:5, 53-60` **Vulnerability Type**: Third-party package and software supply-chain risk **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 invoking this skill, use dlazy minimax-h3 -h for help."}} ``` From `SKILL.md:53-60`: ```markdown You can install on demand without persisting a global binary by running: ```bash npx @dlazy/cli@1.2.3 <command> ``` Or, if you prefer a global install, the skill's `metadata.clawdbot.install` field declares the exact pinned version (`npm install -g @dlazy/cli@1.2.3`). Review the GitHub source before installing. ``` The same package installation and execution mechanisms appear in `SKILL-cn.md:5, 53-60`. ### Technical Analysis The Skill contains no local executable implementation. Instead, it directs the agent to retrieve and execute `@dlazy/cli@1.2.3` from the npm ecosystem through either `npx` or a global npm installation. Pinning the direct package to version `1.2.3` is a useful control against unintended direct-version updates, but it does not independently verify package integrity, provenance, install-time scripts, or the integrity of transitive dependencies. The audited project does not include a lockfile, trusted package hash, signature, vendored source, or locally reviewable implementation that would establish that the downloaded npm artifact matches the referenced repository. Both installation methods may execute package ...[truncated 2226 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Vendor the required CLI implementation into a reviewed source tree or provide a minimal local client whose full behavior can be audited with the Skill. 2. Generate and commit a lockfile that fixes the complete transitive dependency graph, not only the direct package version. 3. Verify the downloaded package using a trusted cryptographic digest, signed provenance, or registry signature before execution. 4. Ensure that the verified npm artifact is reproducibly linked to the referenced source repository and release tag. 5. Disable npm lifecycle scripts where they are unnecessary, for example with `--ignore-scripts`, after confirming that the package functions safely without them. 6. Avoid global installation. Run the CLI inside an isolated environment with a dedicated cache and no write access outside required output directories. 7. Restrict filesystem access to explicitly selected input files and output paths. 8. Expose only the required API credential and environment variables to the subprocess. 9. Apply outbound network controls permitting only the documented service endpoints. 10. Review the package and its complete dependency graph before every approved version change, and automate dependency vulnerability and provenance checks in the release process. ]]>
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (1)

Intent-Code Divergence

Low
Confidence
91% confidence
Finding
The manifest and surrounding documentation describe this skill as a video-generation capability, but the sample JSON output declares an output item with `type: "image"`, `result.png`, and `image/png`. This is an active documentation contradiction that can mislead an agent about the actual artifact type it should expect and handle.

Static analysis

No suspicious patterns detected.