🖼️ Photo — 照片

v1.0.0

Video Ai：用此技能将照片和视频转为动态影像，支持 JPG、PNG、MP4、MOV，最大 500MB。社交媒体创作者用它把…

0· 17·0 当前·0 累计

by @mory128

开发工具代码生成文件处理 AI模型访问 CI/CD

下载技能包

最后更新

2026/4/19

安全扫描

VirusTotal

无害

查看报告

OpenClaw

可疑

medium confidence

The skill's stated purpose (cloud photo/video rendering) matches most of what it requests, but the runtime instructions expect the agent to read local install/config paths and the SKILL.md frontmatter despite the registry metadata not listing those paths — this mismatch and implicit filesystem probing is worth caution before installing.

评估建议

This skill appears to do what it says: call a remote service to render videos and it needs a single API token (NEMO_TOKEN). Before installing, consider: 1) The SKILL.md instructs the agent to detect install paths (~/.clawhub/, ~/.cursor/skills/) and read YAML frontmatter — that means the agent may probe some local paths; make sure you are comfortable with the agent accessing those locations. 2) Only provide a NEMO_TOKEN scoped to the minimum necessary privileges (or let the skill use its anonymo...

详细分析 ▾

ℹ 用途与能力

The skill's name and description (convert photos/videos to videos) line up with requiring a NEMO_TOKEN and calling a remote rendering API. However, the SKILL.md frontmatter declares a required config path (~/.config/nemovideo/) and the running instructions expect to detect install paths (~/.clawhub/, ~/.cursor/skills/) and to read the file's YAML frontmatter at runtime. The registry metadata provided to the evaluator did not list any required config paths — that inconsistency suggests the skill's declared capabilities/requirements in the catalog are out-of-sync with the runtime instructions.

⚠ 指令范围

Instructions explicitly tell the agent to: read NEMO_TOKEN from the environment (expected); generate an anonymous token from the vendor endpoint if no env token exists (expected); create sessions, upload files (multipart uploads or URLs), poll render status, and include custom attribution headers. They also instruct detecting the agent's install path (~/.clawhub/ or ~/.cursor/skills/) and reading the SKILL.md YAML frontmatter at runtime to populate X-Skill-Version. Detecting/install-paths and reading local files is filesystem access beyond the simple API integration and is not declared in the registry metadata. If an agent has access to arbitrary local paths, these instructions could cause it to probe the filesystem for those locations.

✓ 安装机制

No install spec and no code files — instruction-only skill. This minimizes write-to-disk and supply-chain risk; nothing is being downloaded or installed by the skill itself.

ℹ 凭证需求

The skill only requires one credential (NEMO_TOKEN) as primaryEnv, which is proportionate to calling a third-party render API. The instructions also create an anonymous token if NEMO_TOKEN is not present (POST to vendor endpoint). There are no additional unrelated secrets requested. Users should verify what privileges the NEMO_TOKEN grants on the vendor side (account access, billing, long-lived operations).

✓ 持久化与权限

always:false and normal autonomous invocation settings. The skill does not request to be forced-always-enabled and does not provide install scripts or attempt to modify other skills or system-wide settings in the provided instructions.

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

Photo Video AI 1.0.0 — Initial Release - Convert photos and videos (JPG, PNG, MP4, MOV, up to 500MB) into animated, shareable 1080p MP4 videos in 30-60 seconds using cloud GPUs. - Simple workflow for uploading, editing, previewing, and exporting video slideshows with transitions, music, and text overlays. - Built-in session management with automatic anonymous login and 100 free credits for new users (valid 7 days). - Supports aspect ratio selection, BGM, and multiple export formats; optimized for quick edits and social media sharing. - Includes user-friendly error handling, status updates, and step-by-step instructions for all major video creation tasks. - All processing and rendering handled securely on remote servers—no local installation required.

● 无害

安装命令

点击复制

官方npx clawhub@latest install photo-video-ai

镜像加速npx clawhub@latest install photo-video-ai --registry https://cn.longxiaskill.com镜像同步中

需要定制？告诉我你的需求 →

技能文档

快速开始

随时待命。把照片/视频拖进来，或描述你想做的效果。

试试说：

“把五张产品图和一段背景音乐转成 1080p MP4”
“把我的照片做成带转场和音乐的平滑幻灯片视频”
“把照片合集变成社交媒体创作者用的可分享视频”

连接后端

处理请求前，先连后端 API，显示“连接中…”。 若环境已有 NEMO_TOKEN，直接用它建会话；否则领取免费试用 token：

生成 UUID 作为 client ID
POST 到 https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token，带 X-Client-Id 头
返回含 100 积分、7 天有效的 token，用作 NEMO_TOKEN

接着创建会话： POST https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent Bearer 授权，body {"task_name":"project","language":"en"} 返回的 session_id 后续每次请求都要用。告诉用户“已就绪”，技术细节不展示。

# Photo Video AI — 把照片变成可分享的视频把照片/视频发给我，并描述想要的效果。AI 视频在远程 GPU 节点渲染，本机无需安装。示例：上传五张产品图+背景音乐，输入“做成带转场和音乐的幻灯片”，约 30-60 秒返回 1080p MP4，全程云端完成。注意：≤10 张照片可保持处理快、视频精简。

输入与动作映射

用户提到 photo video ai、比例、文字层、音轨等，经关键词与意图分类路由：

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

云端渲染流程

导出任务排队到云 GPU，合成视频层、按平台压缩（H.264，最大 1080×1920），30-90 秒内返下载链接。 session token 带渲染 job ID，关闭页面前请等待，否则任务孤立。

所有请求必须带： Authorization: Bearer X-Skill-Source X-Skill-Version X-Skill-Platform 缺失将 402 失败。

运行时读取 YAML 头：

X-Skill-Source: photo-video-ai
X-Skill-Version: 取头信息 version
X-Skill-Platform: 按安装路径检测（~/.clawhub/ → clawhub，~/.cursor/skills/ → cursor，其他 → unknown）

API 基址：https://mega-api-prod.nemovideo.ai 创建会话：POST /api/tasks/me/with-session/nemo_agent body {"task_name":"project","language":""} → 返 task_id, session_id

发消息（SSE）：POST /run_sse body {"app_name":"nemo_agent","user_id":"me","session_id":"","new_message":{"parts":[{"text":""}]}} Accept: text/event-stream，最大 15 分钟超时

上传：POST /api/upload-video/nemo_agent/me/ 文件：multipart -F "files=@/path" 或 URL：{"urls":[""],"source_type":"url"}

积分：GET /api/credits/balance/simple → 返 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，不扣积分）：POST /api/render/proxy/lambda body {"id":"render_","sessionId":"","draft":,"output":{"format":"mp4","quality":"high"}} 轮询 GET /api/render/proxy/lambda/ 每 30 秒，至 status = completed，下载地址在 output.url

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

错误码

| 码 | 含义 | 处理 | |-----|-------|--------| | 0 | 成功 | 继续 | | 1001 | token 失效/过期 | 重新鉴权 |

数据来源：ClawHub ↗ · 中文优化：龙虾技能库