← 返回 Skills 市场
foxwzh

seaart video

作者 FoxWzh · GitHub ↗ · v1.0.3 · MIT-0
cross-platform ✓ 安全检测通过
124
总下载
0
收藏
0
当前安装
4
版本数
在 OpenClaw 中安装
/install seaart-video
功能描述
Use this skill to generate high-quality AI videos using the SeaArt platform (seaart.ai). Supports both Text-to-Video and Image-to-Video generation using vari...
安全使用建议
This skill appears to do what it says: it uses your SeaArt 'T' cookie to call seaart.ai endpoints via an included Python script. Before installing or saving your token: 1) Only set SEAART_TOKEN if you trust this skill and the environment where the agent stores config; the token is equivalent to a session cookie. 2) Prefer using a throwaway SeaArt account or revocable token if you are cautious. 3) If you do persist the token, ensure your agent config file is accessible only to you. 4) You can avoid persistence by providing the token each session instead of running /update-config. 5) Review the included script (scripts/generate.py) yourself — it appears to call only seaart.ai endpoints and does not write files, but verifying the full file is recommended. 6) Note minor metadata inconsistencies (the registry shows 'Required env vars: [object Object]' and SKILL.md path assumptions) — these are likely authoring bugs, not indicators of malicious behavior.
功能分析
Type: OpenClaw Skill Name: seaart-video Version: 1.0.3 The skill is a legitimate integration for generating AI videos via the SeaArt platform. It provides clear instructions for users to securely provide a session token (SEAART_TOKEN) and uses a transparent Python script (scripts/generate.py) to communicate with official seaart.ai endpoints. The code follows security best practices by scoping cookies to the correct domain, avoiding shell interpolation, and explicitly documenting external network calls and data handling.
能力评估
Purpose & Capability
The skill claims to generate videos via SeaArt and only requires a SeaArt session token (T cookie) plus Python and requests; those needs match the stated purpose. Model IDs, API endpoints, and parameters in the manifest and script align with video generation.
Instruction Scope
SKILL.md confines actions to checking for SEAART_TOKEN, collecting prompt/model/aspect data, and running the included Python script which contacts only seaart.ai endpoints. It asks users to set the token via /update-config and explicitly warns not to paste the token in chat. Minor issues: SKILL.md references running the script from ~/.claude/skills/... (an assumed install path) and the top-level registry metadata shows a broken 'Required env vars: [object Object]' entry — both are minor inconsistencies but not malicious.
Install Mechanism
There is no installer that downloads arbitrary code; the skill is instruction-only with an included Python script. The script uses the requests library (pip3), which is expected for HTTP API calls. No external or unusual download URLs or extract steps are present.
Credentials
Only one credential is requested (SEAART_TOKEN) and it is the correct type for authenticating to seaart.ai. The requested binaries (python3/pip3) are reasonable. No unrelated secrets or config paths are requested.
Persistence & Privilege
The skill does not have always:true. It instructs users to persist the SEAART_TOKEN in the agent config via /update-config; storing the token locally is reasonable for convenience but means the agent (if allowed to invoke skills autonomously) could call the SeaArt API without the user re-entering the token. This is expected behavior for API-integrations but worth noting for users who want stricter control.
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 seaart-video
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /seaart-video 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
v1.0.3
- Clarified the required environment variables and binaries in the metadata section, including descriptions for each. - Added explicit requirement and description for pip3, needed to install the requests library if not already present. - No functional or code changes; documentation improvements only.
v1.0.2
- No user-facing changes in this release. - Version incremented without detection of file changes.
v1.0.1
- Improved security instructions for storing the SEAART_TOKEN: now users are directed to never send tokens via chat and to use the /update-config command in their terminal. - Updated documentation to remove advice for Claude Code and OpenClaw environments, simplifying the setup flow. - Clarified privacy and local data handling in the Security & Privacy section—emphasizing secure token storage and no data logging. - Adjusted sample usage commands and references to third-party models for accuracy. - No code changes; this update affects documentation only.
v1.0.0
SeaArt Video Generator v1.0.0 — Initial Release - Enables high-quality AI video generation using the SeaArt platform with Text-to-Video and Image-to-Video support. - Supports multiple generation models including SeaArt Sono Cast (default), Vidu Q3 Turbo, Kling 3.0, and Wan 2.6. - Allows choice of aspect ratios for text prompts (16:9, 9:16, 1:1, 4:3, 3:4). - Provides clear instructions for obtaining and securely storing the required SeaArt authentication token. - Includes guidance on invoking the video generation process and presenting output video URLs.
元数据
Slug seaart-video
版本 1.0.3
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 4
常见问题

seaart video 是什么?

Use this skill to generate high-quality AI videos using the SeaArt platform (seaart.ai). Supports both Text-to-Video and Image-to-Video generation using vari... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 124 次。

如何安装 seaart video?

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

seaart video 是免费的吗?

是的,seaart video 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

seaart video 支持哪些平台?

seaart video 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 seaart video?

由 FoxWzh(@foxwzh)开发并维护,当前版本 v1.0.3。

💬 留言讨论