[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-81338":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":13,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":17,"stars90d":15,"forks30d":15,"starsTrendScore":18,"compositeScore":19,"rankGlobal":10,"rankLanguage":10,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":15,"starSnapshotCount":15,"syncStatus":16,"lastSyncTime":33,"discoverSource":34},81338,"promptlab-image-video-to-prompt","gracech0322-cmd\u002Fpromptlab-image-video-to-prompt","gracech0322-cmd","PromptLab is a free and open-source image and video prompt generator. It turns local videos and web\u002Flocal images into ready-to-use prompts, with Video to Prompt output generated in a Seedance 2.0-style prompt format.","",null,"TypeScript",45,1,42,0,2,3,6,46.7,"MIT License",false,"main",true,[25,26,27,28,29],"ai-prompt-generator","image-to-prompt","seedance2","seedance2-prompt","video-to-prompt","2026-06-12 04:01:33","# PromptLab: Image & Video to Prompt\n\nTurn local videos, web\u002Flocal images, and short ideas into ready-to-use AI prompts.\n\nPromptLab is a Chrome extension for AI creators, prompt learners, visual reference study, and short video prompt reverse engineering. It helps you turn local video files, web images, local image files, and short creative ideas into clean prompts you can copy and use.\n\n## 🎬 Table of Contents\n\n- [What Is PromptLab?](#what-is-promptlab)\n- [What Can PromptLab Do?](#what-can-promptlab-do)\n- [Installation](#installation)\n- [How to Get Free Gemini API Key?](#how-to-get-free-gemini-api-key)\n- [API Key Privacy](#api-key-privacy)\n- [Settings](#settings)\n- [Video Frame Sampling](#video-frame-sampling)\n- [Image Prompt Logic](#image-prompt-logic)\n- [Video Prompt Logic](#video-prompt-logic)\n- [Current Limitations](#current-limitations)\n- [Privacy](#privacy)\n- [Roadmap](#roadmap)\n- [License](#license)\n\n## What Is PromptLab?\n\nPromptLab is a Chrome extension that turns local videos, web images, local image files, and short ideas into ready-to-use AI prompts. It is made for AI creators, prompt learners, visual reference study, and short video prompt reverse engineering.\n\nPromptLab does not currently support full online video analysis from platforms like Instagram, TikTok, X, Facebook, or YouTube.\n\nFor video analysis, please upload a local video file.\n\n## What Can PromptLab Do?\n\n### Local Video to Prompt\n\n![Local video to prompt](assets\u002Flocal-video-to-prompt.jpg)\n\nUpload a local video file. PromptLab samples key frames and turns them into one clean Final Result prompt.\n\n- Best for short local videos under 60 seconds\n- Extracts key frames automatically\n- Designed for short creative videos, AI videos, ads, reels, and cinematic clips\n- Default frame sampling mode: Standard\n- Supports Fast \u002F Standard \u002F Detailed frame sampling modes\n\n### Web Image to Prompt\n\n![Web image to prompt](assets\u002Fweb-image-to-prompt.jpg)\n\nRight-click a web image. PromptLab reads the image and creates a ready-to-use image prompt.\n\n- Uses a dedicated image prompt template\n- Outputs one clean Final Result prompt\n- Optimized for image generation and visual recreation\n\n### Local Image to Prompt\n\n![Local image to prompt](assets\u002Flocal-image-to-prompt.jpg)\n\nUpload a local image file. PromptLab uses an image-focused template to create one final prompt for image generation.\n\n- Uses the same image-oriented output style as web image analysis\n- Outputs one clean Final Result prompt\n- Built for image generation and visual recreation\n\n### Prompt Enhancer\n\n![Prompt enhancer](assets\u002Fprompt-enhancer.jpg)\n\nTurn a short idea into a fuller prompt that is easier to copy and use.\n\n- Includes a Video Prompt mode and an Image Prompt mode\n- Video Prompt expands a short idea into a Seedance 2.0-style video prompt\n- Image Prompt expands a short idea into a general image generation prompt\n- Uses one clean Final Result output\n- Works as an optional helper tool alongside the main image and video workflows\n\n### Prompt History\n\n- Automatically saves recent generation records\n- Keeps up to 20 recent records\n\n### Simple Output\n\n- One unified result: Final Result\n- No long analysis report\n- No unnecessary labels like \"Image Prompt:\" or \"Analysis:\"\n- Copy and use directly\n\n## Installation\n\n### Option 1: Install from Download ZIP, recommended for most users\n\n1. Click Code -> Download ZIP on GitHub.\n2. Unzip the file.\n3. Open Chrome and go to `chrome:\u002F\u002Fextensions\u002F`.\n4. Turn on Developer mode.\n5. Click Load unpacked.\n6. Select the `dist\u002F` folder.\n\nImportant:\n\n- Select `dist\u002F`\n- Do not select the project root folder\n- Do not select `src\u002F`\n- Do not select `assets\u002F`\n\nIf there is no `dist\u002F` folder in the downloaded ZIP, use Option 2 to build it manually.\n\nAfter the extension loads, open Settings and add your Gemini API Key.\n\n### Option 2: Build manually\n\n1. Clone or download this repository.\n2. Open Terminal in the project folder.\n3. Run:\n\n```bash\nnpm install\nnpm run build\n```\n\n4. After build finishes, open Chrome and go to `chrome:\u002F\u002Fextensions\u002F`.\n5. Turn on Developer mode.\n6. Click Load unpacked.\n7. Select the generated `dist\u002F` folder.\n\nAfter the extension loads, open Settings and add your Gemini API Key.\n\n## How to Get Free Gemini API Key?\n\nPromptLab uses your own Gemini API Key.\n\nYou can get a Gemini API Key from Google AI Studio:\n\nhttps:\u002F\u002Faistudio.google.com\u002F\n\nAfter you install the extension, open Settings and paste your Gemini API Key.\n\n## API Key Privacy\n\nYour Gemini API Key is stored locally in your browser extension storage.\n\nThe project developer cannot see, collect, or access your API Key.\n\nImages, video frames, and prompts may be sent to Gemini for analysis based on your Gemini API usage.\n\nIf you are worried about API usage or accidental exposure, you can set a daily API usage limit in Google AI Studio or Gemini settings.\n\n## Settings\n\n### Gemini API Key\n\nRequired. Used to send image or video frame analysis requests to Gemini.\n\n### Frame Sampling Mode\n\nThis setting only applies to local video analysis.\n\nDefault mode: Standard\n\nAvailable modes:\n\n- Fast\n- Standard\n- Detailed\n\nFast:\n\n- Fewer frames\n- Faster analysis\n- Best for quick prompt generation\n\nStandard:\n\n- Default recommended mode\n- Balanced speed and prompt quality\n- Best for most local videos\n\nDetailed:\n\n- More frames\n- Better for complex motion, transitions, ads, and AI showcase videos\n- Slower and may use more API resources\n\nImage analysis does not use frame sampling settings.\n\n## Video Frame Sampling\n\nFast:\n\n- 5 frames\n- 0%, 25%, 50%, 75%, 95%\n\nStandard:\n\n- duration \u003C= 10s: 6 frames\n- 10s \u003C duration \u003C= 30s: 10 frames\n- 30s \u003C duration \u003C= 60s: 14 frames\n- duration > 60s: 16 frames\n\nDetailed:\n\n- duration \u003C= 10s: 10 frames\n- 10s \u003C duration \u003C= 30s: 16 frames\n- 30s \u003C duration \u003C= 60s: 24 frames\n- duration > 60s: 32 frames\n\nPromptLab samples from 0% to 95% of the video duration. It avoids 100% to reduce the chance of black frames or seek errors.\n\n## Image Prompt Logic\n\nImage analysis uses its own image-oriented prompt template.\n\nIt focuses on:\n\n- Subject\n- Scene\n- Composition\n- Style\n- Lighting\n- Color palette\n- Mood\n- Texture\n- Material\n- Perspective\n- Visual quality\n- Image generation keywords\n\n## Video Prompt Logic\n\nVideo analysis uses a video-oriented prompt template.\n\nThe final result is written in a **Seedance 2.0-style prompt format**, focusing on clear subject description, action details, camera language, motion continuity, and cinematic scene structure.\n\nAlthough the output is designed around Seedance 2.0-style prompting, it can also be used as a strong reference prompt for other AI video generation models. You may adjust the wording based on the model you use.\n\n**Read More**:[Seedance 2.0 Prompt Library](https:\u002F\u002Fgithub.com\u002Fgracech0322-cmd\u002Fseedance-2-prompt-library)\n\nIt focuses on:\n\n- Subject\n- Scene\n- Action\n- Camera language\n- Motion\n- Pacing\n- Visual continuity\n- Cinematic structure\n- AI video generation style\n\n## Current Limitations\n\nPromptLab does not support:\n\n- Instagram video extraction\n- TikTok video extraction\n- X \u002F Twitter video extraction\n- Facebook video extraction\n- YouTube video extraction\n- DRM or protected video extraction\n- m3u8 video downloading\n- Streaming video parsing\n\nFor best video results, upload a local video file.\n\n## Privacy\n\nPromptLab uses your own Gemini API Key.\n\nYour API Key is stored locally in browser extension storage.\n\nThe developer cannot access your API Key.\n\nDepending on your Gemini API usage, images, video frames, and prompts may be sent to Gemini for analysis.\n\nDo not analyze private, sensitive, or copyrighted content unless you have permission.\n\nYou can set a daily API usage limit in Google AI Studio.\n\n## Roadmap\n\n- [x] Local video to prompt\n- [x] Web image to prompt\n- [x] Local image to prompt\n- [x] Fast \u002F Standard \u002F Detailed video frame sampling\n- [x] Prompt history, up to 20 records\n\nPlanned:\n\n- [ ] Better image source fallback\n- [ ] Prompt style presets\n- [ ] Batch image analysis\n- [ ] Direct video URL import\n- [ ] Optional online video visible segment capture\n\n## License\n\nThis project is licensed under the MIT License.\n\nThe PromptLab name, logo, and branding assets are not included in the MIT License and may not be used without permission.\n","PromptLab 是一个免费且开源的图像和视频提示生成器，能够将本地视频、网页或本地图片转换为可直接使用的AI提示。其核心功能包括从本地视频中提取关键帧并生成Seedance 2.0风格的提示文本，以及从网页或本地图片中创建用于图像生成的提示。此外，它还提供了一个提示增强工具，可以将简短的想法扩展成更详细的提示内容。该工具主要采用TypeScript编写，并以Chrome扩展的形式存在，适用于AI创作者、提示学习者、视觉参考研究以及短视频提示反向工程等场景。","2026-06-11 04:04:41","CREATED_QUERY"]