← 返回 Skills 市场
Opus Clip
作者
whitejohnk-26
· GitHub ↗
· v1.0.0
· MIT-0
67
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install opus-clip
功能描述
Get short viral clips ready to post, without touching a single slider. Upload your long-form video (MP4, MOV, AVI, WebM, up to 500MB), say something like "cl...
安全使用建议
This skill appears to be what it says: a cloud-based video clipper that requires a single service token. Before installing or using it, consider: (1) The service host (mega-api-prod.nemovideo.ai) will receive any uploaded video — do not upload sensitive content unless you trust the service and reviewed its privacy policy/retention rules. (2) If NEMO_TOKEN is not provided, the skill will POST to an anonymous-token endpoint to obtain a short-lived token automatically (this is normal but is a network call you should expect). (3) The SKILL.md asks the agent to detect its install path to set an X-Skill-Platform header — that can reveal which client/runtime path is in use; if you prefer not to expose that, provide a token and/or review how the agent builds headers. (4) There is a small metadata inconsistency (frontmatter mentions a config path while the registry metadata did not); not necessarily malicious but worth verifying. Lastly, the skill source/homepage is unknown — if you need stronger assurance, ask the publisher for a homepage or privacy/security documentation before uploading production or sensitive videos.
功能分析
Type: OpenClaw Skill
Name: opus-clip
Version: 1.0.0
The skill provides a functional interface for the Opus Clip video processing service, communicating with the backend at mega-api-prod.nemovideo.ai. It handles authentication via environment variables or anonymous token generation, manages video upload/rendering sessions, and processes server-sent events (SSE) for real-time updates. All identified behaviors, including platform detection for attribution and local configuration access, are consistent with the stated purpose of a video editing integration and show no signs of malicious intent or data exfiltration.
能力评估
Purpose & Capability
Name/description (cloud video clipper) matches the declared primary credential (NEMO_TOKEN) and the SKILL.md instructs only calls to nemovideo.ai endpoints for uploads, SSE, rendering and exports. Nothing else (no AWS keys, no unrelated service tokens) is requested. Note: the SKILL.md frontmatter lists a configPaths entry (~/.config/nemovideo/) even though the registry summary above reported no required config paths — a discrepancy in metadata but not itself a functional mismatch.
Instruction Scope
The instructions keep scope focused on uploading videos, starting sessions, streaming SSE, polling render status, and exporting results from mega-api-prod.nemovideo.ai. They instruct the agent to use NEMO_TOKEN (or obtain a short-lived anonymous token via the anonymous-token endpoint) and to include attribution headers. One privacy-relevant instruction: detect the agent's install path to set X-Skill-Platform (e.g., check for ~/.clawhub/ or ~/.cursor/skills/), which implies filesystem inspection to derive a header value. Otherwise the runtime steps do not ask for unrelated files or other environment variables.
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is downloaded or written to disk by an installer. This is low risk from an installation perspective.
Credentials
Only one credential is required (NEMO_TOKEN), which is appropriate for calling the vendor API. The skill also describes a local config path in its frontmatter (~/.config/nemovideo/), but the registry's requirement list did not include config paths — this mismatch is a metadata inconsistency. The SKILL.md will attempt to mint an anonymous token if NEMO_TOKEN is absent, which is expected behavior but means the agent will perform a network POST to obtain a token.
Persistence & Privilege
The skill does not request always: true or other elevated platform privileges and does not instruct modifying other skills or system-wide settings. It uses session tokens for transient jobs; no persistent system presence is requested.
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 opus-clip - 安装完成后,直接呼叫该 Skill 的名称或使用
/opus-clip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
v1.0.0
Opus Clip 1.0.0 — Initial Release
- Instantly convert long-form videos into short, viral-ready clips with AI-powered editing—no manual sliders or timelines required.
- Simple upload and prompt-based workflow: describe your desired output (e.g., "clip the best 60-second moments and add captions for TikTok") and receive a ready-to-share 1080p MP4.
- Supports video uploads up to 500MB in popular formats (MP4, MOV, AVI, WebM) with fast cloud processing.
- Integrated session management, credits system, and robust error handling for a smooth user experience.
- Designed for content creators and marketers seeking to repurpose long videos into engaging short clips quickly and easily.
元数据
常见问题
Opus Clip 是什么?
Get short viral clips ready to post, without touching a single slider. Upload your long-form video (MP4, MOV, AVI, WebM, up to 500MB), say something like "cl... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 67 次。
如何安装 Opus Clip?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install opus-clip」即可一键安装,无需额外配置。
Opus Clip 是免费的吗?
是的,Opus Clip 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Opus Clip 支持哪些平台?
Opus Clip 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 Opus Clip?
由 whitejohnk-26(@whitejohnk-26)开发并维护,当前版本 v1.0.0。
推荐 Skills