← 返回 Skills 市场
mory128

Video Jiggle

作者 mory128 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
115
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install video-jiggle
功能描述
stabilize video clips into stabilized video clips with this skill. Works with MP4, MOV, AVI, WebM files up to 500MB. content creators use it for removing cam...
安全使用建议
This skill appears to do what it says (upload your video, run cloud stabilization, return MP4). Before installing, consider: 1) The backend domain (mega-api-prod.nemovideo.ai) is not documented with a public homepage or privacy/terms links — you should confirm you trust that service before uploading private videos. 2) Although it declares NEMO_TOKEN as required, the skill will obtain an anonymous token itself if none is provided; if you want to avoid the skill contacting external endpoints automatically, do not install it or block its network access. 3) The SKILL.md implies the agent may check local install paths (~/.clawhub/, ~/.cursor/) to set headers — if you are uncomfortable with any filesystem inspection, do not install. 4) Prefer supplying your own account token (NEMO_TOKEN) from a known, trusted provider rather than relying on anonymous tokens. 5) Ask the publisher for a homepage, privacy policy, and source code if you need stronger assurance; absence of those increases risk. If you proceed, avoid uploading sensitive or private video content until you verify the service and its data retention policies.
功能分析
Type: OpenClaw Skill Name: video-jiggle Version: 1.0.0 The skill facilitates video stabilization by uploading local files to a remote API (mega-api-prod.nemovideo.ai) and using a Server-Sent Events (SSE) stream for command processing, which are high-risk capabilities. It also includes instructions for the agent to hide technical details from the user and perform environment fingerprinting by checking its installation path to determine the platform (SKILL.md). While these behaviors are plausibly needed for the stated purpose, the combination of broad file/network access and reduced transparency warrants a suspicious classification.
能力评估
Purpose & Capability
The skill claims to stabilize video via a cloud API and the SKILL.md contains API endpoints and upload/export flows consistent with that. Requesting a single API credential (NEMO_TOKEN) is reasonable for this purpose. However, the SKILL.md also includes a configPaths metadata entry (~/.config/nemovideo/) while the registry metadata listed no required config paths — this mismatch is an inconsistency. Also the skill can operate without NEMO_TOKEN by acquiring an anonymous token itself, so declaring NEMO_TOKEN as strictly required is misleading.
Instruction Scope
Instructions are largely scoped to creating sessions, uploading files, streaming SSE, polling renders, and returning download URLs — all expected for a cloud render service. Points to note: the skill instructs deriving an 'X-Skill-Platform' header by detecting install paths (e.g., checking for ~/.clawhub/ or ~/.cursor/skills/), which implies reading the filesystem to infer environment; this is not strictly necessary for video stabilization and expands the agent's I/O footprint. The skill also tells the agent how to request an anonymous token and to generate UUIDs for X-Client-Id, meaning it will contact an external service automatically if no NEMO_TOKEN is present.
Install Mechanism
This is an instruction-only skill with no install spec and no code files — lowest-risk install mechanism because nothing is written to disk by an installer. Behavior is entirely runtime network calls described in SKILL.md.
Credentials
Only one environment variable is declared (NEMO_TOKEN), which is proportional for an API-backed service. However, because the skill will request an anonymous token itself when NEMO_TOKEN is absent, marking NEMO_TOKEN as required is inconsistent. The skill does not request unrelated cloud credentials, but it does imply reading install/config paths and deriving headers from local install paths, which increases its scope of access to local environment information.
Persistence & Privilege
The skill is not marked always:true and doesn't request persistent system-wide privileges. It does allow autonomous invocation (platform default) which increases its runtime blast radius only if combined with other red flags; here there are no install-time persistence mechanisms or cross-skill configuration changes.
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 video-jiggle
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /video-jiggle 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
v1.0.0
- Initial release of Video Jiggle, an AI-powered video stabilization skill. - Supports MP4, MOV, AVI, and WebM files up to 500MB with 1080p MP4 export. - Removes camera shake from handheld or action footage in 20–40 seconds via cloud GPU processing. - Handles all connection, session, and rendering workflows automatically—no installation required. - User-friendly: just upload your clip and describe the desired result. - Includes error handling, session management, and common workflow tips for smooth usage.
元数据
Slug video-jiggle
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Video Jiggle 是什么?

stabilize video clips into stabilized video clips with this skill. Works with MP4, MOV, AVI, WebM files up to 500MB. content creators use it for removing cam... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 115 次。

如何安装 Video Jiggle?

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

Video Jiggle 是免费的吗?

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

Video Jiggle 支持哪些平台?

Video Jiggle 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 Video Jiggle?

由 mory128(@mory128)开发并维护,当前版本 v1.0.0。

💬 留言讨论