← 返回 Skills 市场
Miniature Build Video
作者
parallel world
· GitHub ↗
· v1.0.5
· MIT-0
119
总下载
0
收藏
0
当前安装
6
版本数
在 OpenClaw 中安装
/install miniature-build-video
功能描述
Generate vertical miniature build & reveal shorts (WeryAI): text or finished-shot image to shallow-depth push-ins, lights coming on, immersive camera. Use wh...
安全使用建议
This package is coherent with its stated purpose, but review and consider the following before installing: (1) Only set WERYAI_API_KEY if you trust WeryAI and the skill author — the key is used to authenticate API calls and uploads. (2) Do not include secrets in prompts or image metadata. (3) Prefer supplying public https image URLs; if you supply local file paths the script will read those files and upload them to WeryAI (consuming credits). (4) Use node scripts/video_gen.js --dry-run first to inspect the JSON shape without performing network calls. (5) If you need strict data control, do not set the key or run this in an isolated/test environment. Review scripts/video_gen.js yourself if you have any doubt.
功能分析
Type: OpenClaw Skill
Name: miniature-build-video
Version: 1.0.5
The skill provides video generation via the WeryAI API and includes a Node.js script (scripts/video_gen.js) that can read local files and upload them to remote endpoints (api.weryai.com and api-growth-agent.weryai.com). While these capabilities are plausibly needed for the stated purpose of image-to-video generation and the documentation (SKILL.md) includes security warnings and instructions for the agent to obtain user consent before file access, the inherent risk of arbitrary file exfiltration via the script's upload functionality meets the threshold for a suspicious classification.
能力评估
Purpose & Capability
Name/description (miniature video generation) matches the actual footprint: Node.js CLI + network calls to WeryAI. Required binary (node) and required env var (WERYAI_API_KEY) are appropriate and expected.
Instruction Scope
SKILL.md and resources instruct the agent to run scripts/video_gen.js and to provide either public https image URLs or local image paths; the bundled script will read local image files (if given) and upload them to WeryAI. The README warns about this and recommends review/consent and using --dry-run first. This behavior is in-scope for an image→video tool but is important to notice because local files will be read and transmitted to the provider if used.
Install Mechanism
Instruction-only skill with a bundled script; no installer or remote downloads. Nothing is fetched or executed from untrusted URLs during install.
Credentials
Only one secret (WERYAI_API_KEY) is required and is the expected credential for the declared provider. No unrelated credentials, host overrides, or excessive env requirements are present.
Persistence & Privilege
always: false and normal autonomous invocation settings. The skill does not request persistent system-wide privileges or modify other skills’ configuration.
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 miniature-build-video - 安装完成后，直接呼叫该 Skill 的名称或使用
/miniature-build-video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
v1.0.5
**Summary: Improved clarity, tightened API & env var requirements, and clarified prompt and security handling.**
- Added detailed requirements for `WERYAI_API_KEY` as the single necessary runtime secret; moved env info to frontmatter and OpenClaw metadata.
- Added `resources/WERYAI_VIDEO_API.md` for explicit documentation of API fields and CLI commands.
- Improved documentation: clarified prompt expansion process, usage of public image URLs, and user confirmation before generation.
- Detailed security guidance—always review included scripts, especially regarding local file/image uploads.
- Made API hostnames in `video_gen.js` fixed and explicit (cannot override via env vars).
- Restructured and condensed usage, prerequisites, and risk disclosure sections for concise registration and onboarding.
v1.0.4
No file or documentation changes detected in this version.
- No updates or changes from the previous version (1.0.0).
- All features, parameters, and usage remain the same.
v1.0.3
- No file or content changes detected in this version.
- Version number updated, but SKILL.md content remains unchanged from the previous release.
v1.0.2
No changes detected in this version.
- Version number updated to 1.0.2, but no file or documentation changes found.
- No user-facing feature, bugfix, or documentation modifications.
v1.0.1
No user-facing changes detected for version 1.0.1.
- No update in features, documentation, or system requirements.
- No modifications to files or SKILL.md content.
v1.0.0
Initial release – generate immersive miniature build and showcase videos from prompts or model images.
- Generates short videos (5s default, 10/15s for build process) with cinematic macro effects, extreme shallow depth of field, and rich lighting.
- Supports both scene-text prompts (e.g., "赛博朋克工作室，霓虹风") and direct use of model photos for camera movement/lighting animation.
- Includes ambient audio (ASMR, environmental sounds) for increased immersion.
- Exposes easy configuration for scene type, mood, and presentation focus.
- Works via Node.js script; requires only WERYAI API key.
- No dependency on other Cursor skills.
元数据
常见问题
Miniature Build Video 是什么？
Generate vertical miniature build & reveal shorts (WeryAI): text or finished-shot image to shallow-depth push-ins, lights coming on, immersive camera. Use wh... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 119 次。
如何安装 Miniature Build Video？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install miniature-build-video」即可一键安装，无需额外配置。
Miniature Build Video 是免费的吗？
是的，Miniature Build Video 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Miniature Build Video 支持哪些平台？
Miniature Build 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 Miniature Build Video？
由 parallel world（@zoucdr）开发并维护，当前版本 v1.0.5。
推荐 Skills