首页 / 技能 / Ai — 人工智能

💬 Ai — 人工智能

v1.0.0

Wechat Video 只需输入需求，即可将 60 秒产品演示视频一键转为 1080p、适配微信分享的视频。无论转换还是格式调整……

0· 15·0 当前·0 累计
mhogan2013-9 头像by @mhogan2013-9
开发工具通信工具视频处理系统工具数据库
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
可疑
medium confidence
该技能的声明目的（格式化微信视频）与其描述的 API 调用大体一致，但存在内部不一致及若干指令级细节不符，安装或提供凭据前需谨慎。
评估建议
This skill mostly does what it says — it talks to a Nemovideo-like backend to upload and render short videos — but a few things are inconsistent and worth checking before you install or hand over credentials: - Only provide a NEMO_TOKEN if you trust the backend/service. That token can be used to upload files, start renders, and consume credits on your account. If possible, test with an anonymous token (the SKILL.md shows how to obtain one) rather than your primary account token. - Ask the publi...
详细分析 ▾
ℹ 用途与能力
The declared purpose (cloud-based video formatting for WeChat) aligns with the API endpoints and operations documented (upload, render, export). Requesting a single service token (NEMO_TOKEN) is proportional for a cloud backend. However, the SKILL.md frontmatter lists a required config path (~/.config/nemovideo/) while the registry metadata earlier listed no required config paths — this mismatch is inconsistent.
⚠ 指令范围
The runtime instructions direct the agent to obtain or reuse a bearer token, create sessions, upload videos (multipart or URL), stream SSE responses, and poll render status — all expected for this service. Concerns: (1) the file-upload examples explicitly show local file paths (files=@/path) which could imply reading arbitrary local paths if the agent has file access; the instructions don't clearly constrain uploads to user-supplied attachments. (2) The header 'X-Skill-Platform' is supposed to be auto-detected from install path — instruction-only skills often cannot reliably detect an install path, so this requirement is unrealistic. (3) The SKILL.md frontmatter requiring a config directory (contradicting registry metadata) suggests either sloppy packaging or an unexplained need to read local config.
✓ 安装机制
No install spec and no code files (instruction-only) — lowest-risk delivery mechanism. Nothing is downloaded or written to disk by an installer.
ℹ 凭证需求
The skill only requests one credential: NEMO_TOKEN (primaryEnv). That is proportionate for a cloud render service. Important caveat: that single token can grant access to the user's account on the backend (jobs, uploads, credits), so it's sensitive — the SKILL.md also documents steps to obtain an anonymous token if not provided, which is safer for trial use. The apparent frontmatter request for a local config path is inconsistent with the declared requirements and should be clarified.
✓ 持久化与权限
always is false and the skill is instruction-only; it does not request permanent inclusion or system-wide configuration changes. Autonomous invocation is allowed by default (normal) but does increase runtime scope; this is not combined with elevated privileges in this package.
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

- AI WeChat Video v1.0.0 首次发布。 - 通过简单聊天指令，即时将 60 秒产品演示视频格式化并导出为 1080p MP4，可直接用于微信。 - 无缝后端连接：使用环境 token 自动建立会话，或匿名访问并赠送免费额度。 - 云端 GPU 视频处理支持裁剪、字幕、画幅比例及多种输出格式（如 mp4、mov、jpg）。 - 清晰的错误处理与会话管理；向用户展示连接、额度及导出状态。 - 专为微信视频快速优化设计，无需时间线编辑或导出设置。

● 无害

安装命令

点击复制
官方npx clawhub@latest install ai-wechat-video
镜像加速npx clawhub@latest install ai-wechat-video --registry https://cn.longxiaskill.com

技能文档

快速开始

有视频片段？发给我并告诉我需求——AI 视频优化交给我。

试试说：

  • “把 60 秒产品演示转成 1080p MP4”
  • “给这条视频加字幕、剪成方形，适配朋友圈”
  • “帮微信创作者转换并格式化视频”

连接后端

处理请求前先连 API，显示“连接中…”。 若环境变量已有 NEMO_TOKEN，直接建会话；否则：
  • 生成 UUID 作为 client ID
  • POST https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token，带 X-Client-Id
  • 返回含 100 积分、7 天有效的 token，用作 NEMO_TOKEN

接着建会话： POST https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent Bearer 鉴权，body {"task_name":"project","language":"en"} 返回的 session_id 后续均需使用。 告诉用户已就绪，技术细节不展示。

# AI WeChat Video — Format and Export WeChat Videos 把视频拖进聊天框，告诉我需求。云端 GPU 完成 AI 优化，本地无需安装。 典型流程：发 60 秒产品演示 → 要求“朋友圈方形字幕版” → 30-60 秒后得 1080p MP4 下载链接。 注意：时长≤1 分钟，符合朋友圈限制。

意图映射

含 ai wechat video、比例、文字、音轨等关键词的提示，按关键字路由：

| 用户说法 | 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/发文件” | → §3.2 上传 | ✅ | | 其余（生成、剪辑、加 BGM…） | → §3.1 SSE | ❌ |

云端渲染流程

导出任务排队到 GPU 节点，合成层、平台压缩（H.264，最大 1080×1920），30-90 秒返下载链接。 会话 token 携带任务 ID，提前关标签页会导致任务孤立。

所有请求必须带： Authorization: Bearer X-Skill-Source: ai-wechat-video X-Skill-Version: 1.0.0 X-Skill-Platform: clawhub/cursor/unknown 缺失返回 402。

API 基址：https://mega-api-prod.nemovideo.ai

  • 建会话：POST /api/tasks/me/with-session/nemo_agent
  • 发消息（SSE）：POST /run_sse，Accept: text/event-stream，最长 15 min
  • 上传：POST /api/upload-video/nemo_agent/me/，支持文件 multipart 或 URL JSON
  • 积分：GET /api/credits/balance/simple
  • 状态：GET /api/state/nemo_agent/me//latest
  • 导出（免费）：POST /api/render/proxy/lambda，轮询同一地址至 status=completed，得 output.url

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

数据来源：ClawHub ↗ · 中文优化：龙虾技能库