Back to skill

Security audit

文生图 Seedream 5.0 Lite

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed wrapper for a cloud image-generation CLI, with ordinary SaaS credential, upload, and install considerations but no artifact-backed malicious behavior.

Install this only if you trust dLazy and the @dlazy/cli npm package. Prefer npx or another isolated execution method over a global install, review the referenced CLI source when practical, and only pass prompts or local files you are comfortable uploading to dLazy's service.

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 Executes with User Privileges## Vulnerability Details **File Location**: `SKILL.md:5, 56-62`; `SKILL-cn.md:5, 56-62` **Vulnerability Type**: T08: Insecure Dependencies **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 seedream-5.0-lite -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":"When this skill is invoked, dlazy seedream-5.0-lite -h may be used to view help."}} ``` ```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 externally distributed npm package `@dlazy/cli`. The package is pinned to version `1.2.3`, which prevents automatic selection of a later version, but the Skill does not provide or verify a cryptographic integrity digest for the retrieved package. The dependency implementation is also absent from the audited project, so its lifecycle scripts and runtime behavior cannot be validate ...[truncated 2158 chars]
Remediation
## Remediation Suggestions 1. Vendor a reviewed copy of the required client implementation into the Skill or provide a minimal auditable client whose source is included in the project. 2. Verify the exact npm artifact with a cryptographic integrity value obtained through a trusted lockfile or controlled artifact manifest. 3. Use a private or internally mirrored package registry after performing dependency and malware scanning. 4. Review the package's direct and transitive dependencies, published artifact contents, lifecycle scripts, and correspondence with the referenced source repository. 5. Avoid global installation. Prefer an isolated, non-privileged environment with a dedicated cache and restricted filesystem access. 6. Disable npm lifecycle scripts with `--ignore-scripts` when the verified package does not require them. 7. Run the CLI in a sandbox that permits access only to explicitly selected input and output files and only to the documented network endpoints. 8. Keep the API credential scoped to the minimum required permissions, protect the configuration file with user-only permissions, and support prompt credential rotation and revocation. 9. Add automated software-composition analysis, package-signature or provenance verification, and recurring dependency review to the release process.
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
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
94% confidence
Finding
The trigger keywords are overly broad and include common terms like '豆包', 'seedream', and generic image-generation phrases, which can overlap with ordinary user conversation and cause unintended skill activation. In an agent setting, accidental invocation can lead to unplanned external API calls, upload of referenced local files, and unnecessary consumption of user credits.

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
94% confidence
Finding
The listed triggers include generic terms such as "doubao," "seedream," and especially broad phrases like "generate image" and "text to image," which can overlap with ordinary user requests rather than clearly scoping when this specific skill should activate. The section also lacks negative examples or constraints to distinguish when the skill should not be invoked.

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.

Natural-Language Policy Violations

Low
Confidence
85% confidence
Finding
该文件整体以中文编写并向智能体下达中文交互指令,但未说明这是用户可选语言,或仅在用户选择中文时适用。按规则,若技能强制特定语言而没有用户 opt-in,属于语言/locale 政策风险。

Natural-Language Policy Violations

Low
Confidence
76% confidence
Finding
The description and headings present both English and Chinese content inline, which can impose a locale/language presentation choice without stating that the agent should follow the user's preferred language. Because the policy requires avoiding forced language behavior unless opt-in or clearly justified, this should be made explicit.

Static analysis

No suspicious patterns detected.