← 返回 Skills 市场
Weryai Video Generator
作者
WeryAI-Developer
· GitHub ↗
· v0.1.2
· MIT-0
193
总下载
0
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install weryai-video-generator
功能描述
Generate and transform WeryAI videos from text, images, storyboard frames, or first-frame and last-frame guidance. Use when the user needs text-to-video, ima...
安全使用建议
This skill generally looks coherent for WeryAI video generation, but check the following before installing or running it:
- Packaging: The scripts import many modules from a '../../../../../core/...' path that are not included in this bundle. That means these scripts likely expect a larger monorepo or shared runtime. If you plan to run them standalone, they will fail — ask the author for a complete package or the required core files.
- Secrets: Only WERYAI_API_KEY is required. Keep that API key secret and do not commit it into the repository. Prefer exporting it into the runtime environment.
- Data exfiltration surface: The skill will upload local media referenced in requests to WeryAI (via the upload endpoint). Do not submit sensitive local files unless you trust WeryAI and want them uploaded.
- Network endpoints: Defaults point to api.weryai.com and api-growth-agent.weryai.com. Do not override WERYAI_BASE_URL or WERYAI_MODELS_BASE_URL to untrusted hosts.
- First run: Use the provided dry-run commands (node scripts/models-video.js and node scripts/wait-video.js --dry-run) to verify configuration without spending credits. Also verify you have account credits before submitting paid jobs.
If you need higher assurance, request the missing core modules or a packaged release (or a vetted npm package) and review the core code referenced by the re-exports before running in production.
功能分析
Type: OpenClaw Skill
Name: weryai-video-generator
Version: 0.1.2
The skill bundle is a legitimate integration for the WeryAI video generation service. It provides scripts for text-to-video, image-to-video, and status tracking, with clear instructions in SKILL.md and AGENTS.md to guide the AI agent on parameter mapping and bounded polling to prevent infinite loops. The code follows standard practices for API wrappers, includes dry-run capabilities for safety, and lacks any indicators of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
The name, description, required binary (node), and required env var (WERYAI_API_KEY) align with a video-generation integration. However, many script files re-export implementations from '../../../../../core/weryai-core' and '../../../../../core/weryai-video' which are not present in this package manifest; this suggests the package is intended to run inside a larger monorepo or shared runtime. That packaging mismatch is an operational inconsistency (not necessarily malicious) and will cause runtime failures if executed standalone.
Instruction Scope
SKILL.md and AGENTS.md instruct the agent to submit jobs, poll until completion, optionally upload local media via the API, and return playable video URLs — all coherent with the stated purpose. Important scope notes: local files referenced by the user are uploaded to the WeryAI upload endpoint (i.e., user local media will be sent to the external API), and the runtime enforces bounded polling (timeouts). There are no instructions to read unrelated local files or other environment secrets beyond the declared API key.
Install Mechanism
No install spec or remote downloads are present; the skill is node-script based and expects Node >=18 on PATH. There are no suspicious external URLs or archive extracts. The only operational caveat is the missing 'core' modules referenced by the scripts (packaging/integration issue), not an installation risk.
Credentials
The skill requires a single primary credential (WERYAI_API_KEY) and optionally allows overrides for base URLs and timeout-related env vars. This is proportionate to a third-party API integration. No unrelated secrets or broad credential sets are requested.
Persistence & Privilege
The skill does not request 'always: true', does not modify other skills, and uses normal agent invocation defaults. There is no evidence it attempts to persist beyond its intended scope or escalate privileges.
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 weryai-video-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/weryai-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
**Major refactor to improve agent usability and remove vendored dependencies.**
- All 25 vendored core and video utility scripts have been removed for a cleaner codebase.
- Documentation clarified to emphasize result-first, bounded-wait workflows (auto-polling with timeouts).
- Usage guidance now stresses example prompts and natural language instructions (not raw shell commands).
- Default behaviors and safety confirmations are unchanged for end users.
- Security, authentication, and parameter selection guidance remain, with improved clarity.
v0.1.1
Version 0.1.1 of weryai-video-generator
- Added support for media input from local files or non-http(s) sources, with auto-upload prior to video generation.
- Default configuration now enables `generate_audio=true` for audio-capable models.
- New scripts and modules for media uploading and improved video generation workflows.
- Updated documentation for environment variable setup, quick verification, and local media handling.
v0.1.0
weryai-video-generator 0.1.0
- Initial release of the official skill for generating WeryAI videos from text, images, storyboard frames, or guided transitions.
- Supports text-to-video, image-to-video, first-frame/last-frame video, storyboard-to-video, model switching, and video task status checks.
- Defaults to SEEDANCE_2_0 model with 5s, 720p, 9:16 aspect ratio, and no audio, while allowing customizable parameters.
- Provides dry-run payload previews, trusted API host checks, and model capability lookups for safer paid runs.
- Includes operator-style user guidance, intent routing, and progressive parameter mapping for common creative requests.
元数据
常见问题
Weryai Video Generator 是什么？
Generate and transform WeryAI videos from text, images, storyboard frames, or first-frame and last-frame guidance. Use when the user needs text-to-video, ima... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 193 次。
如何安装 Weryai Video Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install weryai-video-generator」即可一键安装，无需额外配置。
Weryai Video Generator 是免费的吗？
是的，Weryai Video Generator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Weryai Video Generator 支持哪些平台？
Weryai 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 Weryai Video Generator？
由 WeryAI-Developer（@weryai-developer）开发并维护，当前版本 v0.1.2。
推荐 Skills