360 Secure SkillHub
360 安全技能中心

Video Maker Free

核心功能:依托AI自动完成素材排版、照片动效、配乐、配音、转场等全流程编辑,一键生成专业视频。典型场景:房产/产品照片转宣传视频、产品文本转讲解视频、零散家庭视频片段转故事视频等。输入支持多类素材组合,可通过API配置风格、导出格式等参数;输出多格式无水印高清视频,适配各平台发布需求。

peand-rover
peand-rover
热度 67.72
最近更新 2026-05-11 17:36:59
分类:创作依赖:第三方 API应用场景:视频编辑 / 社交媒体 / 故事创作 / 图像生成 ...

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add video-maker-free
Metadata
name
video-maker-free
version
1.2.1
displayName
Video Maker Free — Make Videos from Photos Text and Clips with AI for Free
description
Make videos for free using AI — combine photos, text, and video clips into polished content with transitions, music, voiceover, subtitles, and effects. NemoVideo creates videos from whatever you have: a set of product photos becomes a promotional video, a text script becomes a narrated explainer, scattered phone clips become a cohesive story, and raw ideas become publishable content — all free, all full quality, no watermarks.
metadata
{"openclaw":{"emoji":"🎬","requires":{"env":[],"configPaths":["~/.config/nemovideo/"]},"primaryEnv":"NEMO_TOKEN"}}
homepage
https://nemovideo.com
repository
https://github.com/nemovideo/nemovideo_skills

Video Maker Free — Make Any Video from Photos, Text or Clips

Most people who need a video don't start with footage — they start with whatever they have. A real estate agent has 15 property photos. A small business owner has a product description. A student has presentation slides. A parent has scattered phone clips from a birthday party. A marketing team has bullet points from a strategy meeting. None of these are "footage" in the traditional sense, but every one of them could be a compelling video. The gap between "what I have" and "a finished video" is the editing process: importing assets into software, arranging them on a timeline, adding motion to photos (Ken Burns, pan-and-zoom), timing text overlays, finding and mixing music, recording or generating voiceover, adding transitions between elements, and exporting at the right settings for each platform. NemoVideo bridges that gap with one command. Provide whatever you have — photos, text, clips, or a combination — and describe the video you want. The AI assembles, animates, narrates, scores, and exports a finished video. Photos get motion and transitions. Text becomes narrated scenes with supporting visuals. Clips get trimmed, color-matched, and joined. The output is a real video — not a slideshow with a filter, but a produced piece of content with professional pacing, audio, and visual quality.

Use Cases

  1. Photos → Product Video (30-60s) — An Etsy seller has 8 product photos and needs a video for Instagram. NemoVideo: sequences photos with smooth Ken Burns motion (slow zoom on detail shots, pan across wide shots), adds product name and price as animated text overlays, underlays upbeat acoustic music, applies a consistent warm color grade, and exports 9:16 for Instagram and 1:1 for the listing page. Eight static images become a dynamic product showcase.
  2. Text → Explainer Video (60-180s) — A SaaS startup has a 300-word product description and needs a landing page video. NemoVideo: breaks the text into Problem → Solution → Benefits → CTA scenes, generates supporting visuals for each scene (office frustration, clean dashboard UI, happy team, pricing page), narrates with a professional voice, adds animated statistics, and exports 16:9 for the website. No filming, no stock footage budget.
  3. Mixed Media → Story Video (2-5 min) — A parent has 20 phone photos and 8 short clips from their child's first birthday party. NemoVideo: sorts by timestamp, sequences photos with gentle motion and clips at key moments (cake smash, candle blowing, gift opening), adds cheerful background music, overlays the child's name and age as animated titles, and exports as a shareable family video with a clean opening and closing.
  4. Slides → Training Video (5-15 min) — An HR department has a 30-slide presentation that nobody reads. NemoVideo: converts each slide into a video scene with animated bullet points, generates voiceover narration from the slide notes, adds transitions between topics, inserts knowledge-check pause points, and exports as a training module that employees actually watch. Slide decks become engaging video content.
  5. Bullet Points → Social Content (15-30s per video) — A marketing manager has 10 product features as bullet points and needs 10 short social videos. NemoVideo batch-generates: each bullet becomes a 15-second video with bold animated text, supporting visual, music, and CTA. Ten social videos from ten lines of text — a month of daily posts produced in one batch.

How It Works

Step 1 — Provide Your Materials

Upload photos, video clips, text, or any combination. NemoVideo accepts all formats and intelligently assembles mixed media.

Step 2 — Describe the Video

Tell NemoVideo what you want: the story, the style, the mood, the platform. Detailed instructions or "make something beautiful from these photos" — both work.

Step 3 — Generate

curl -X POST https://mega-api-prod.nemovideo.ai/api/v1/generate \
  -H "Authorization: Bearer $NEMO_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "skill": "video-maker-free",
    "prompt": "Make a 45-second product showcase video from 8 product photos. Style: clean and modern with white background accents. Each photo gets 4-5 seconds with smooth Ken Burns motion (slow zoom on details, pan on wide shots). Add product name as animated text: Artisan Ceramic Mug Collection. Price: from $28. Music: warm acoustic guitar at -14dB. Color grade: bright and clean. End frame: Shop now at artisanceramics.com. Export 9:16 for Instagram and 1:1 for website.",
    "media_type": "photos",
    "photo_count": 8,
    "style": "clean-modern",
    "music": "acoustic-guitar-warm",
    "music_volume": "-14dB",
    "text_overlays": ["Artisan Ceramic Mug Collection", "From $28"],
    "cta": "Shop now at artisanceramics.com",
    "exports": ["9:16", "1:1"],
    "watermark": false
  }'

Step 4 — Preview and Share

Preview. Adjust photo order, transition timing, text placement, or music. Export and share — free, full quality.

Parameters

Parameter Type Required Description
prompt string ✅ Describe the video and materials
media_type string "photos", "clips", "text", "mixed"
style string "clean-modern", "cinematic", "playful", "elegant", "bold"
music string "acoustic", "lo-fi", "corporate", "cinematic", "electronic"
music_volume string "-12dB" to "-22dB"
voice string Voiceover: "warm-male", "friendly-female", "none"
text_overlays array Text to display as animated overlays
cta string Call-to-action text
photo_motion string "ken-burns", "parallax", "slide", "zoom"
duration string "30 sec", "45 sec", "60 sec", "natural"
exports array ["16:9", "9:16", "1:1"]
batch array Multiple videos from separate material sets
watermark boolean Always false

Output Example

{
  "job_id": "vmf-20260328-001",
  "status": "completed",
  "source_materials": "8 photos",
  "watermark": false,
  "outputs": [
    {
      "format": "9:16",
      "resolution": "1080x1920",
      "duration": "0:44",
      "file_size_mb": 12.4,
      "photo_motion": "ken-burns (zoom + pan)",
      "text_overlays": 3,
      "music": "acoustic-guitar-warm at -14dB"
    },
    {
      "format": "1:1",
      "resolution": "1080x1080",
      "duration": "0:44",
      "file_size_mb": 11.8
    }
  ]
}

Tips

  1. Photos with motion beat static slideshows — Ken Burns pan-and-zoom adds life to still images. A slow zoom into a product detail feels cinematic. A gentle pan across a room feels like a camera movement. Static photos displayed full-frame feel like PowerPoint.
  2. 4-5 seconds per photo is the engagement sweet spot — Shorter than 3 seconds feels rushed and the viewer can't absorb the image. Longer than 6 seconds and attention drifts. 4-5 seconds with smooth motion holds attention perfectly.
  3. Music without speech can be louder — Photo/product videos without voiceover benefit from music at -12 to -14dB (louder than the -18dB used under speech). The music carries the emotional energy that speech would normally provide.
  4. Batch generation scales content instantly — 10 products × 1 video each = 10 social media posts. Batch-process with consistent style for brand cohesion but unique content per product.
  5. Multi-format export from one generation — 9:16 for Instagram/TikTok + 1:1 for feed + 16:9 for website. Three formats, one command, one consistent video.

Output Formats

Format Resolution Use Case
MP4 9:16 1080x1920 Instagram / TikTok / Stories
MP4 16:9 1920x1080 YouTube / website / email
MP4 1:1 1080x1080 Instagram feed / Twitter
GIF 720p Preview / thumbnail

Related Skills