← 返回 Skills 市场
Seedance 2 Video Generator
作者
Weiwei Fan
· GitHub ↗
· v0.1.2
· MIT-0
204
总下载
1
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install seedance-2-video-generator
功能描述
Generate Werydance 2.0 videos through WeryAI for text-to-video, image-to-video, multi-image video, and first-frame/last-frame transitions. Use when you need...
安全使用建议
This skill appears to do what it says: it will call WeryAI APIs and upload any media files you provide. Before installing or invoking it: (1) only provide local file paths you trust—the skill will read and upload those files to WeryAI; (2) ensure you are comfortable giving your WERYAI_API_KEY (check billing/permissions on that key); (3) note the skill defaults to adding audio unless you explicitly request silence; and (4) confirm paid submissions when prompted (SKILL.md requests explicit confirmation). If you need to prevent accidental data exfiltration, avoid supplying arbitrary local paths or run in dry-run mode first to see what would be uploaded.
功能分析
Type: OpenClaw Skill
Name: seedance-2-video-generator
Version: 0.1.2
The skill is a legitimate and well-structured API wrapper for the WeryAI video generation service (Werydance 2.0). It facilitates complex video generation workflows including text-to-video, multi-image storyboarding, and 'almighty' mixed-reference modes. While the skill includes high-risk capabilities such as reading and uploading local files to a remote API (scripts/vendor/weryai-core/upload.js), it incorporates robust security safeguards including a strict file extension whitelist, size limits (10MB for images, 50MB for media), and a security warning mechanism that triggers if files are sent to non-official domains. The instructions in SKILL.md and AGENTS.md are strictly aligned with the stated purpose of prompt engineering and workflow management.
能力评估
Purpose & Capability
Skill name/description, required binary (node), and the single required env var (WERYAI_API_KEY) align with the included code, which builds and posts requests to WeryAI endpoints (text-to-video, image-to-video, multi-image-to-video, almighty-reference-to-video). There are no unrelated credentials or unrelated external services requested.
Instruction Scope
SKILL.md and the scripts constrain behavior to generating videos and uploading media. The runtime will stat/read local files you provide (to validate and upload them) and will upload them to the WeryAI upload endpoint; this is expected for a media upload skill but means any local path you pass will be read and transmitted. The skill also defaults to adding audio unless the user explicitly requests silence and requires explicit confirmation before paid submission per SKILL.md.
Install Mechanism
There is no install spec (instruction-only install), and the package includes the JS runtime files. No external download URLs, package installers, or archive extraction are present—risk from installation is low. The runtime uses only standard Node APIs and internal helper modules included in the bundle.
Credentials
The skill declares a single primary credential WERYAI_API_KEY, which is appropriate for a client of the WeryAI API. The code also honors optional WERYAI_* environment overrides (e.g., WERYAI_BASE_URL, poll timeouts) — these are reasonable. No unrelated secrets or high-privilege env vars are requested.
Persistence & Privilege
The skill is not forced-always (always:false), is user-invocable, and does not request elevated agent-wide privileges or modify other skills. It will execute its included scripts when invoked but does not persist beyond that.
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-video-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/seedance-2-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
v0.1.2
**Werydance 2.0 video generator now supports almighty reference modes and local file upload.**
- Added support for multimodal "almighty" reference-to-video generation, accepting images, videos, audios, and text in one request.
- Supports local file paths for images, videos, and audios with automatic uploading—no need for public URLs.
- New strict validations for almighty mode: total media/file count, duration, field limits per OpenAPI.
- Smart routing: selects the correct WeryAI API route based on input types (legacy modes, or almighty mode).
- Dry-run improvements: displays which local files would be uploaded before running paid generation.
- Documentation updates for new modes, constraints, and field guidelines.
v0.1.1
**Seedance-2-video-generator v0.1.1 Changelog**
- Added support for "almighty" mixed-reference video generation (using images, videos, audios, and prompt together).
- Enabled direct uploading of local images, videos, and audios before generation; local files are now automatically uploaded as needed.
- Introduced new almighty API route (`/v1/generation/almighty-reference-to-video`) with input validation for mixed media payloads.
- Updated prompt, input, and confirmation workflows to reflect new modal logic and constraints (e.g., max 500-char prompt, media count limits).
- Documentation and examples improved: covers almighty reference flow, CLI updates, and quickstart for all input modes.
- Backwards compatibility maintained for legacy field aliases and dry-run behavior.
v0.1.0
Initial release of Seedance 2.0 video generator skill for WeryAI.
- Generate Seedance 2.0 videos via WeryAI: supports text-to-video, image-to-video, multi-image video, and first-frame/last-frame transitions.
- Strict about security and input: requires only WERYAI_API_KEY, and media must be public HTTPS URLs.
- Includes prompt expansion and confirmation workflow to ensure production-quality prompts before paid runs.
- Offers compatibility for migrations from WaveSpeed Seedance workflows (image + last_image).
- Provides clear setup, authentication, and usage instructions, with support for dry-run and model capability checks.
- Declares all requirements, dependencies, and modes in metadata for OpenClaw and manual usage.
元数据
常见问题
Seedance 2 Video Generator 是什么？
Generate Werydance 2.0 videos through WeryAI for text-to-video, image-to-video, multi-image video, and first-frame/last-frame transitions. Use when you need... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 204 次。
如何安装 Seedance 2 Video Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install seedance-2-video-generator」即可一键安装，无需额外配置。
Seedance 2 Video Generator 是免费的吗？
是的，Seedance 2 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 Video Generator 支持哪些平台？
Seedance 2 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 Video Generator？
由 Weiwei Fan（@fwwdn）开发并维护，当前版本 v0.1.2。
推荐 Skills