← 返回 Skills 市场
275
总下载
2
收藏
0
当前安装
13
版本数
在 OpenClaw 中安装
/install seedance-2-ai-video-generator
功能描述
Generates video via Loova Seedance 2.0 API (Seedance 2.0 video). Requires LOOVA_API_KEY from .env or environment (get API key at loova.ai). Use when the user...
安全使用建议
This skill appears to do what it says: it needs your LOOVA_API_KEY and will send prompts, URLs, and any local files you provide to api.loova.ai to generate videos (generation can take hours and the script polls once per minute). Before installing: (1) Only provide an API key you trust Loova with and check your account permissions/rate limits; (2) avoid passing paths to sensitive local files — the script will open and upload any file paths you provide; (3) verify which exact task-status endpoint the live API expects (docs include a small /api vs /v1 mismatch); (4) note the script also checks an undocumented LOOAI_API_KEY env var as a fallback — if you rely on strict env control, be aware of this alternate name. If those points are acceptable, the skill is coherent with its stated purpose.
功能分析
Type: OpenClaw Skill
Name: seedance-2-ai-video-generator
Version: 1.0.13
The skill bundle is a legitimate tool for generating AI videos via the Loova Seedance 2.0 API. The core logic in `scripts/run_seedance.py` correctly handles API authentication via environment variables, supports both multipart file uploads and JSON payloads, and implements a standard polling mechanism for long-running tasks. No evidence of data exfiltration, malicious execution, or prompt injection was found; the code and instructions in `SKILL.md` are strictly aligned with the stated purpose of video generation.
能力评估
Purpose & Capability
Name/description, SKILL.md, README, reference, and the included Python script all align: the code posts to Loova Seedance endpoints and polls for a task result. Required env var LOOVA_API_KEY is appropriate for this API and no unrelated services/binaries are requested.
Instruction Scope
Instructions ask the agent to accept user uploads (or URLs), save local uploads in the workspace (e.g., workspace/assets/) and pass file paths to the script. The script opens and uploads any file paths it is given, and polls once per minute for up to ~3 hours. This behavior is expected for a file-upload video generation skill, but be aware that any local file path passed to the script will be opened and transmitted to the external API — do not pass sensitive local files. Minor documentation inconsistencies: some docs reference GET /api/v1/tasks while others show /v1/tasks (missing /api); these are likely typos but worth verifying against the live API.
Install Mechanism
No install spec; included code uses only Python packages listed in requirements.txt (requests, python-dotenv). No downloads from third-party URLs or archive extraction. This is low-risk and proportionate to the functionality.
Credentials
Declared primary credential is LOOVA_API_KEY which matches the script's Authorization header usage. The script also accepts a fallback env var LOOAI_API_KEY (not declared in metadata), a minor discrepancy. The skill does not request additional secrets or system config paths.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent or elevated platform privileges or alter other skills. It will only act when invoked and does not install background services.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install seedance-2-ai-video-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/seedance-2-ai-video-generator触发 - 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
v1.0.13
- No changes detected in this version.
- The skill version and documentation remain the same as the previous release.
v1.0.12
No user-facing changes detected in this release.
- No changes to functionality, documentation, or code.
- Version increment only; content remains the same.
v1.0.11
No user-visible changes in this version.
- No file changes or updates detected.
- Functionality, documentation, and configuration remain unchanged from the previous version.
v1.0.10
- No changes detected in this version.
- Functionality, documentation, and configuration remain unchanged.
v1.0.9
- No changes detected in this version; the skill remains unchanged from the previous release.
v1.0.8
- Removed `curl` from required system binaries; now no system binaries are needed.
- Clarified instructions: for multipart uploads, allow the HTTP client to set the `Content-Type` header automatically, rather than setting it manually.
- No other user-facing changes.
v1.0.7
- Added OpenClaw metadata with homepage, required binaries (curl), required environment variables, and primary environment variable.
- No changes to functionality or user-facing commands.
v1.0.6
- Updated environment variable specification to use required_env_vars and primary_credential fields.
- Removed legacy credentials section in favor of streamlined environment configuration.
- No changes to API, usage, features, or code functionality.
v1.0.4
- Updated environment variable and credentials documentation for improved clarity and compatibility.
- Credentials and required environment variables are now specified in a more structured format.
- No code or functional changes were made.
v1.0.3
- Added required_env_vars and primary_credential fields to SKILL.md for improved environment variable and credential specification.
- No functional or API changes; documentation and skill metadata only.
v1.0.2
- Added a credentials section to declare LOOVA_API_KEY as the primary credential.
- No behavioral or functionality changes.
- This update improves credential management and documentation clarity.
v1.0.1
- Cleanup: Removed unused scripts (check_seedance.py, seedance_poller.py) from the codebase.
- Documentation update: Clarified that LOOVA_API_KEY is required and must be provided via environment or .env, not hardcoded.
- Added explicit "env: required" section to SKILL.md for clearer setup requirements.
v1.0.0
Initial release of Seedance 2.0 AI Video Generator skill
- Generate AI videos from text, images, video, or audio using the Loova Seedance 2.0 API.
- Supports text-to-video, image-to-video, and multi-modal reference input in a single workflow.
- Configurable options: duration (4–15s), aspect ratio, model (standard/fast), and function mode.
- Handles both remote URLs and user-uploaded media, prioritizing URLs when both are provided.
- Automatic polling for video generation status, with immediate error reporting on failure.
- Requires LOOVA_API_KEY for API access; setup and usage documentation included.
元数据
常见问题
Seedance 2.0 Al Video Generator 是什么？
Generates video via Loova Seedance 2.0 API (Seedance 2.0 video). Requires LOOVA_API_KEY from .env or environment (get API key at loova.ai). Use when the user... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 275 次。
如何安装 Seedance 2.0 Al Video Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install seedance-2-ai-video-generator」即可一键安装，无需额外配置。
Seedance 2.0 Al Video Generator 是免费的吗？
是的，Seedance 2.0 Al Video Generator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Seedance 2.0 Al Video Generator 支持哪些平台？
Seedance 2.0 Al Video Generator 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 Seedance 2.0 Al Video Generator？
由 loova（@hjianfeng）开发并维护，当前版本 v1.0.13。
推荐 Skills