← 返回 Skills 市场
chengzeyi

WaveSpeedAI Seedream 4.5 Image Generation/Editing

作者 C · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
358
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install wavespeed-seedream-45
功能描述
Generate and edit images using ByteDance's Seedream V4.5 model via WaveSpeed AI. Supports text-to-image generation and multi-image editing with custom resolu...
安全使用建议
This SKILL.md looks like a normal WaveSpeed/Seedream client guide, but there are important inconsistencies you should resolve before installing or granting access: (1) The README shows using WAVESPEED_API_KEY, but the skill metadata does not list any required credentials — ask the publisher to add WAVESPEED_API_KEY as a required/primary credential so you know what will be requested. (2) Examples import the 'wavespeed' SDK and call wavespeed.upload(); there is no install spec or declared dependency — confirm that your runtime has a vetted wavespeed SDK (and which version) or request the publisher to provide one. (3) The skill's examples read local files and upload them; if you install this skill, ensure the agent runs in a sandbox or that you understand what local files the agent may access. (4) Verify the skill's source and homepage (none provided in registry). If you cannot verify the publisher or the dependency/install provenance, avoid supplying your API key or running the skill with access to sensitive files. Providing the WAVESPEED_API_KEY should only be done if you trust wavespeed.ai and have validated the skill's origin. Providing the missing metadata/install info or a link to an official SDK would likely change this assessment to benign.
功能分析
Type: OpenClaw Skill Name: wavespeed-seedream-45 Version: 1.0.0 The skill bundle describes a legitimate image generation and editing service using the WaveSpeed AI platform. The `SKILL.md` provides clear instructions and code examples for using the service, and crucially, includes a 'Security Constraints' section that advises users on secure practices, such as validating input URLs and protecting API keys. There is no evidence of prompt injection against the AI agent, data exfiltration, malicious execution, or persistence mechanisms. While the functionality involves external API calls and potentially local file access (for `wavespeed.upload`), these are presented as legitimate features for the stated purpose, and the explicit security warnings mitigate concerns about intentional misuse within the skill's definition.
能力评估
Purpose & Capability
The skill's name and description (image generation/editing via WaveSpeed/Seedream) match the runtime instructions in SKILL.md. However, the SKILL.md expects a WAVESPEED_API_KEY and the wavespeed client library, while the registry metadata declares no required environment variables or dependencies. This mismatch is unexpected: a remote API integration should normally declare its primary credential and/or dependencies.
Instruction Scope
The instructions show example code that uploads local files (wavespeed.upload('/path/to/...')) and uses the wavespeed client to call remote APIs. That implies the agent will need to read local files and perform network requests. The registry metadata does not declare any required config paths or environment variables to support these actions. The SKILL.md also contains guidance about not loading untrusted URLs, but examples explicitly instruct reading local files — clarify whether the agent will have filesystem access and how uploads should be authorized.
Install Mechanism
This is an instruction-only skill with no install spec or code files. The examples import 'wavespeed' and reference a Client class: that implies a dependency (npm package or SDK) which is not declared or provided. Without an install spec, the agent/platform must already have that library available; this is an operational gap and could lead to runtime failures or hidden assumptions about available packages.
Credentials
SKILL.md instructs users to set WAVESPEED_API_KEY and shows using it with a Client, but the registry lists no required env vars and no primary credential. Requesting an API key for a third-party service is reasonable for this skill, but the metadata should declare it. The absence of declared credentials in metadata is a discrepancy that prevents risk reviewers from knowing the credential scope required by the skill.
Persistence & Privilege
The skill is not marked always:true and does not request any platform-level persistent presence. It is user-invocable and allows model invocation (standard defaults). No evidence it modifies other skills or system-wide settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install wavespeed-seedream-45
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /wavespeed-seedream-45 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of wavespeed-seedream-45. - Generate and edit images using ByteDance Seedream V4.5 model via WaveSpeed AI. - Supports text-to-image and multi-image editing with custom resolutions up to 4096x4096. - Enhanced typography rendering for high-quality posters and logos. - Code samples provided for text prompts, image editing, multi-image composition, and error handling. - $0.04 per image (generation or editing). - Includes security and API key usage guidance.
元数据
Slug wavespeed-seedream-45
版本 1.0.0
许可证
累计安装 0
当前安装数 0
历史版本数 1
常见问题

WaveSpeedAI Seedream 4.5 Image Generation/Editing 是什么?

Generate and edit images using ByteDance's Seedream V4.5 model via WaveSpeed AI. Supports text-to-image generation and multi-image editing with custom resolu... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 358 次。

如何安装 WaveSpeedAI Seedream 4.5 Image Generation/Editing?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install wavespeed-seedream-45」即可一键安装,无需额外配置。

WaveSpeedAI Seedream 4.5 Image Generation/Editing 是免费的吗?

是的,WaveSpeedAI Seedream 4.5 Image Generation/Editing 完全免费(开源免费),可自由下载、安装和使用。

WaveSpeedAI Seedream 4.5 Image Generation/Editing 支持哪些平台?

WaveSpeedAI Seedream 4.5 Image Generation/Editing 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 WaveSpeedAI Seedream 4.5 Image Generation/Editing?

由 C(@chengzeyi)开发并维护,当前版本 v1.0.0。

💬 留言讨论