🖥️ Screen — 屏幕

v1.0.0

在线录屏编辑器 无需拖动任何滑块,即可将录屏剪辑成可直接发布的成品。上传录屏文件(MP4、MOV、WebM、AVI,最大 500MB),只需说几句……

0· 18·0 当前·0 累计
下载技能包
最后更新
2026/4/19
0
安全扫描
VirusTotal
无害
查看报告
OpenClaw
安全
medium confidence
该技能的需求和运行时说明大体上与连接 nemo 云 API 的在线视频编辑前端一致,但安装前请留意元数据/说明存在细微不一致及隐私考量。
评估建议
This skill appears to do what it says: it uploads your recordings to a third‑party cloud service (nemovideo) for AI editing and requires a NEMO_TOKEN or will fetch an anonymous starter token. Before installing, consider: 1) Privacy — your raw video/audio will be sent to an external server; confirm you’re comfortable with that and check the provider’s privacy/security policy. 2) Credentials — only provide NEMO_TOKEN if you trust the service; prefer using an ephemeral/anonymous token if possible. ...
详细分析 ▾
用途与能力
The skill requests a single API credential (NEMO_TOKEN) and describes endpoints and workflows for uploading, editing, and exporting videos on a nemo cloud service — this aligns with the declared purpose of cloud AI video editing.
指令范围
SKILL.md is explicit about making network calls to https://mega-api-prod.nemovideo.ai (session creation, SSE chat, upload, export). It also describes deriving headers from the YAML frontmatter and detecting X-Skill-Platform from install path locations, which implies checking the agent's install-directory paths; this is within scope for header population but is a modest additional filesystem probe beyond the stated upload/edit flow.
安装机制
Instruction-only skill with no install spec and no code files — nothing is written to disk by an installer, which is the lowest-risk installation model.
凭证需求
Only NEMO_TOKEN is required which is proportional to calling the nemo API. However, SKILL.md/YAML references a config path (~/.config/nemovideo/) in its frontmatter (metadata) while the registry summary showed no required config paths — this mismatch should be clarified. The skill also supports generating an anonymous token by POSTing a UUID to the service if NEMO_TOKEN is absent (expected but worth noting because it contacts an external auth endpoint).
持久化与权限
always:false and no system-wide config modifications are requested. The skill can be invoked autonomously (platform default) but it does not request elevated or persistent privileges.
安全有层次,运行前请审查代码。

运行时依赖

无特殊依赖

版本

latestv1.0.02026/4/19

Screen Recording Editor Online 初始发布——基于云端、快速的 AI 屏幕录制剪辑工具。 - 上传屏幕录制文件(MP4、MOV、WebM、AVI,最大 500MB),用文字描述所需编辑(如剪掉停顿、添加字幕、插入标题卡)。 - 通过关键词/意图匹配与后端 API 工作流,自动完成裁剪、文字、导出、音频和预览操作。 - 云端 GPU 流水线快速导出高质量 1080p MP4 视频。 - 无需复杂编辑界面;提示词自然映射为视频编辑与导出。 - 支持匿名会话,附赠免费额度或直接消耗 NEMO_TOKEN;完整会话与任务管理。 - 针对上传、会话、额度及导出问题提供友好的错误处理指引。

无害

安装命令

点击复制
官方npx clawhub@latest install screen-recording-editor-online
镜像加速npx clawhub@latest install screen-recording-editor-online --registry https://cn.longxiaskill.com

技能文档

快速开始

上传录屏文件,我立即开始AI剪辑。或直接告诉我你的想法。
试试说:
  • “编辑我的录屏文件”
  • “导出1080p MP4”
  • “剪掉停顿、加字幕,并”

建立连接

处理任何请求前,先连接后端API,显示“连接中…”。 若环境变量存在NEMO_TOKEN,直接使用并创建会话;否则获取免费试用token:
  • 生成UUID作为client ID
  • X-Client-Id头POST到https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token
  • 响应含100积分、7天有效的token——用作NEMO_TOKEN

然后创建会话:Bearer授权POST到https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent,body为{"task_name":"project","language":"en"}。响应中的session_id用于后续请求。告诉用户已就绪,聊天中不展示技术细节。

# Screen Recording Editor Online — 在线剪辑与导出录屏 上传录屏,描述需求,云端渲染后下载结果。例如3分钟软件教程录屏,想剪掉停顿、加字幕、开头加标题卡——后端约1-2分钟生成1080p MP4。 提示:先剪掉空白和长停顿,后续编辑更快更干净。

用户指令匹配动作

用户提及录屏编辑器、比例、文字层、音轨等,通过关键词与意图分类路由:

| 用户说… | 动作 | 跳过SSE? | |-------------|--------|----------| | “export”/“导出”/“download”/“发我视频” | → §3.5 导出 | ✅ | | “credits”/“积分”/“balance”/“余额” | → §3.3 积分 | ✅ | | “status”/“状态”/“show tracks” | → §3.4 状态 | ✅ | | “upload”/“上传”/用户发文件 | → §3.2 上传 | ✅ | | 其他(生成、编辑、加BGM…) | → §3.1 SSE | ❌ |

云端渲染详情

每次导出任务排队到云GPU节点,合成视频层,按平台规格压缩(H.264,最大1080×1920),30-90秒内返回下载链接。会话token携带渲染任务ID,关闭标签页前未完成会导致任务孤儿。所有请求指向https://mega-api-prod.nemovideo.ai

主要端点:

  • 会话POST /api/tasks/me/with-session/nemo_agent,body {"task_name":"project","language":""},返回session_id
  • 聊天(SSE)POST /run_sse,带session_id与消息new_message.parts[0].textAccept: text/event-stream,最长15分钟。
  • 上传POST /api/upload-video/nemo_agent/me/,支持multipart文件或JSON URL。
  • 积分GET /api/credits/balance/simple,返回availablefrozentotal
  • 状态GET /api/state/nemo_agent/me//latest,获取当前草稿与媒体信息。
  • 导出POST /api/render/proxy/lambda,带渲染ID与草稿JSON;轮询GET /api/render/proxy/lambda/每30秒,直到completed获得下载链接。

支持格式:mp4、mov、avi、webm、mkv、jpg、png、gif、webp、mp3、wav、m4a、aac。

请求头取自YAML头信息:X-Skill-Sourcescreen-recording-editor-onlineX-Skill-Version取自version字段,X-Skill-Platform根据安装路径检测(~/.clawhub/clawhub~/.cursor/skills/cursor,否则unknown)。每次请求需带Authorization: Bearer 及所有归因头,缺失将触发402。

草稿JSON使用短键:t轨道,tt轨道类型(0=视频,1=音频,7=文字),sg片段,d时长(毫秒),m元数据。

时间线示例: `` Timeline (3 tracks):

  • Video: city timelapse (0-10s)
  • BGM: Lo-fi (0-10s, 35%)
  • Title: "Urban Dreams" (0-3s)
``

数据来源ClawHub ↗ · 中文优化:龙虾技能库