🎬 Adobe
v1.0.0使用此技能将原始视频素材编辑成精致的剪辑片段。支持MP4、MOV、AVI、MKV文件，最高500MB。内容创作者和营销人员使用它进行...
20· 20·0 当前·0 累计
by 下载技能包
最后更新
2026/4/20
安全扫描
OpenClaw
可疑
medium confidence该技能通常与云视频编辑工作流匹配，但几个不匹配和意外行为（误导性名称、隐藏匿名令牌创建和不一致的配置声明）使其意图或来源不明确。
评估建议
This skill appears to be a cloud-based video editor that talks to nemovideo.ai, but its public metadata and naming are inconsistent (it uses 'Adobe' in the name while calling a different provider). Before installing: 1) Be cautious with sensitive videos — files are uploaded to a third-party service. 2) The skill will automatically create an anonymous NEMO_TOKEN if you don't supply one and will connect to the remote backend on first use; if you prefer control, set your own NEMO_TOKEN rather than ...详细分析 ▾
⚠ 用途与能力
The skill claims the name 'Adobe Video Editing' but its runtime targets nemovideo.ai and uses a NEMO_TOKEN—there is no evidence it is an Adobe product or officially affiliated. The requested credential (NEMO_TOKEN) and API endpoints do align with the described cloud-rendering capability, but the use of the Adobe name is misleading.
ℹ 指令范围
SKILL.md stays within the editing/exporting domain (session creation, SSE, upload, export, credit checks). However it instructs the agent to automatically connect on first open and to silently obtain an anonymous token if none is present, and it instructs hiding raw token/API responses from the user—behaviors that can be surprising and reduce transparency about network activity.
✓ 安装机制
Instruction-only skill with no install spec or code files; nothing is written to disk by an installer. This is lower risk from an installation perspective.
⚠ 凭证需求
Only one credential (NEMO_TOKEN) is required which is proportionate for a remote API, but the SKILL.md frontmatter references a config path (~/.config/nemovideo/) while the registry metadata lists no required config paths—this inconsistency in declared config access is concerning. Also the skill will generate and use anonymous tokens automatically if none are set.
ℹ 持久化与权限
always:false and no install hooks are good. The skill does instruct storing session_id for subsequent requests and the frontmatter references a config directory (possible persistent storage), but it doesn't specify where or how secrets/session IDs are stored. No elevated platform privileges are requested.
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
latestv1.0.02026/4/20
Adobe Video Editing 技能的初始发布。 - 可以编辑原始视频片段（MP4、MOV、AVI、MKV，最高500MB）成为精致的1080p MP4 片段。 - 支持通过prompt上传视频、剪辑、转场、文字覆盖和BGM编辑。 - 快速的云GPU处理（每个片段1-2分钟，较短的片段处理速度更快）。 - 无需本地Adobe软件；所有渲染都发生在服务器端。 - 用户身份验证自动处理（匿名令牌，7天内100次免费credits）。 - 包括交互式状态更新、引导工作流和方便的导出/下载功能。
● 可疑
安装命令
点击复制官方npx clawhub@latest install adobe-video-editing
镜像加速npx clawhub@latest install adobe-video-editing --registry https://cn.longxiaskill.com 镜像可用
技能文档
----------|--------|----------| | "导出" / "导出" / "下载" / "发送给我视频" | → §3.5 导出 | √ | | "积分" / "积分" / "余额" / "余额" | → §3.3 积分 | √ | | "状态" / "状态" / "显示轨道" | → §3.4 状态 | √ | | "上传" / "上传" / 用户发送文件 | → §3.2 上传 | √ | | 其他内容（生成、编辑、添加 BGM...） | → §3.1 SSE | × |
云渲染管道详细信息
每个导出作业都在云 GPU 节点上排队，合成视频层，应用平台规格压缩（H.264，最高 1080x1920），并在 30-90 秒内返回下载 URL。会话令牌携带渲染作业 ID，因此在完成之前关闭标签页会使作业成为孤儿。每个请求都需要三个属性头，并且必须与此文件的前置内容匹配：| 头部 | 值 |
|---|---|
X-Skill-Source | adobe-video-editing |
X-Skill-Version | 前置内容 version |
X-Skill-Platform | 自动检测：clawhub / cursor / unknown 从安装路径 |
Authorization: Bearer 和所有属性头 — 省略它们会在导出时触发 402。
API 基础：https://mega-api-prod.nemovideo.ai
创建会话：POST /api/tasks/me/with-session/nemo_agent — 正文 {"task_name":"project","language":""} — 返回 task_id、session_id。
发送消息（SSE）：POST /run_sse — 正文 {"app_name":"nemo_agent","user_id":"me","session_id":"","new_message":{"parts":[{"text":""}]}}，头部 Accept: text/event-stream。最大超时：15 分钟。
上传：POST /api/upload-video/nemo_agent/me/ — 文件：多部分 -F "files=@/path"，或 URL：{"urls":[""],"source_type":"url"}。
积分：GET /api/credits/balance/simple — 返回 available、frozen、total。
会话状态：GET /api/state/nemo_agent/me//latest — 关键字段：data.state.draft、data.state.video_infos、data.state.generated_media。
导出（免费，无需积分）：POST /api/render/proxy/lambda — 正文 {"id":"render_","sessionId":"","draft":,"output":{"format":"mp4","quality":"high"}}。每 30 秒轮询 GET /api/render/proxy/lambda/，直到 status = completed。下载 URL 在 output.url 中。支持格式：mp4、mov、avi、webm、mkv、jpg、png、gif、webp、mp3、wav、m4a、aac。
读取 SSE 流
文本事件直接发送给用户（在 GUI 翻译后）。工具调用保持内部。心跳和空data: 行意味着 t