在线图片 EXIF 隐私清理 + 报告导出工具
把一批图片拖入网页即可扫描/展示 EXIF 与敏感字段，并一键清理后打包下载。
英文 Prompt:
You are a senior full-stack engineer. Build a production-ready web app: "Online EXIF Cleaner & Privacy Report".
Requirements:
- Users can drag-and-drop multiple images (JPG/PNG/HEIC optional) in the browser.
- Read EXIF data client-side and show a table: camera model, datetime, GPS, orientation, software, custom tags.
- Provide toggles: remove GPS only / remove all data / normalize orientation.
- Output: cleaned images (same format when possible) and a privacy report (JSON + CSV) summarizing removed fields.
- Batch download as ZIP.
Tech stack:
- Frontend: Next.js (App Router) + Type + Tailwind.
- data: exifr (or equivalent) for reading; for writing/stripping data use a safe approach (canvas re-encode for JPEG/PNG) and clearly explain tradeoffs.
- ZIP: fflate or JSZip.
Deliverables (must include all):
- Full source code.
- A clear file tree.
- Local run commands.
- Dockerfile + docker-compose.
- Deployment guide (Vercel for frontend + optional Node server if needed).
- At least 8 tests (unit or e2e) OR a QA checklist with at least 12 items.
Also include:
- Accessibility (keyboard, ARIA), i18n-ready strings.
- Error handling for corrupt images, large batches, and memory limits.
中文释义: 让 AI 生成一个可上线的“EXIF 扫描与清理”在线工具：支持批量导入、展示敏感元数据、按规则清理、导出报告并打包下载；同时给出源码、目录结构、运行/部署命令与测试或 QA 清单。
在线图片批量压缩与质量预设工具
提供多种压缩策略（按质量/按目标体积/按最长边），并在浏览器中预览压缩前后对比与体积变化。
英文 Prompt:
Build "Online Batch Image Compressor" as a complete web app.
Features:
- Batch import images, show a queue with original size, dimensions, and format.
- Compression modes:
1) Quality slider (JPEG/WebP/AVIF when supported)
2) Target max file size (try iterative quality search)
3) Resize by max width/height
- Before/after comparison UI with zoom and split view.
- Preserve or strip data (integrate with EXIF logic).
- Export selected results and download ZIP.
Implementation constraints:
- No server-side processing required; do everything in the browser using modern APIs.
- Provide fallbacks when AVIF is not supported.
Deliverables:
- Complete Next.js + TS code .
- File tree, run commands, Docker deployment.
- Tests or QA checklist (>= 10 cases) focused on quality, size targets, and edge cases.
中文释义: 生成一个“批量压缩”在线工具的完整项目：多种压缩模式、前后对比、批量导出与 ZIP 下载，并附源码/文件树/运行命令/部署与测试或 QA。
在线批量裁剪与缩放（规则预览）工具
按统一规则（居中裁剪/按人像比例/固定边距）对多张图片批量处理，并提供可回滚的预览确认流程。
英文 Prompt:
Create a web app "Batch Crop & Resize with Rule Preview".
Core UX:
- Import many images; choose a processing rule (center-crop, contain, cover, fixed margin, custom aspect ratio).
- Show a preview grid (before/after) and allow per-image overrides.
- Support exporting processed images and a manifest JSON (original -> output mapping).
Engineering:
- Next.js + TS + Tailwind.
- Use canvas for processing.
- Handle EXIF orientation correctly.
Deliverables:
- Full code + file tree.
- Run + build commands.
- Docker + deploy notes.
- Provide at least 5 automated tests AND a QA checklist.
中文释义: 生成一个支持“批量裁剪/缩放 + 规则预览 + 单张覆盖调整 + 导出清单”的在线工具项目，并包含可运行/可部署的完整交付物。
在线图片格式转换器（JPG/PNG/WebP/AVIF）
把现有图片批量转换为目标格式，支持透明通道、背景填充与质量参数，并生成转换报告。
英文 Prompt:
Build "Online Image Format Converter".
Functions:
- Batch convert between JPG, PNG, WebP, and AVIF (when available).
- Options: background color for JPG, keep transparency for PNG/WebP, quality settings.
- Provide a conversion report (CSV) including format, size savings, and any fallbacks used.
Constraints:
- Client-side only processing.
- Clear handling when a browser cannot encode AVIF.
Deliverables:
- Full Next.js TS project, file tree.
- Commands to run/dev/build.
- Docker + deployment guide.
- QA checklist >= 12 items.
中文释义: 生成一个“格式转换”在线工具：批量转换、参数可控、报告可导出、可部署可运行，并带 QA/测试。
在线图片去重与相似度检测（不上传服务器）工具
在本地浏览器计算感知哈希/指纹，找出重复或近似图片，便于清理素材库。
英文 Prompt:
Create "Local-Only Image Duplicate Finder".
Requirements:
- Batch import images and compute perceptual hashes (pHash/aHash/dHash).
- Cluster images by similarity threshold; show groups and allow marking for export.
- Emphasize privacy: do not upload images.
Deliverables:
- Full code (Next.js + TS).
- File tree, run commands, Docker.
- Provide tests (hash correctness) or QA checklist.
中文释义: 生成一个“图片去重”在线工具：本地计算相似度/指纹，分组展示重复图，支持导出与 QA/测试。
在线图片颜色提取与调色板导出工具
从图片中提取主色/辅助色，导出为 CSS 变量、Tailwind config 或 JSON，便于做 UI 主题。
英文 Prompt:
Build "Image Color Palette Extractor".
Features:
- Import images; extract dominant colors and accessible palette suggestions.
- Export formats: CSS variables, Tailwind theme snippet, JSON.
- Provide contrast checks (WCAG) against light/dark backgrounds.
Deliverables:
- Next.js TS app, file tree, commands.
- Docker + deployment guide.
- QA checklist >= 10.
中文释义: 生成一个“颜色提取与导出”在线工具：提取调色板、输出到工程可用配置，并包含部署与 QA。
在线图片批量加边距与画布扩展（非生成）工具
为现有图片批量扩展画布（填充背景色/透明），用于统一尺寸与留白规范。
英文 Prompt:
Create "Batch Canvas Padding Tool".
Behavior:
- Batch import images.
- Choose target canvas size and padding strategy (center, top-left, custom).
- Background fill options: solid color, transparent when possible.
- Export results + manifest JSON.
Deliverables:
- Full Next.js + TS project.
- File tree + run/build commands.
- Docker + deploy guide.
- QA checklist.
中文释义: 生成一个“批量加边距/扩展画布”的在线工具：只处理现有图片，不涉及生成内容；并给出完整可运行交付物与 QA。
在线图片批量命名规范检查与导出清单工具
对导入图片做命名规则校验（前缀/序号/尺寸/语言），生成问题列表与修复建议，并导出清单。
英文 Prompt:
Build "Image Asset Naming Validator".
Requirements:
- Import images; validate filenames against configurable rules (regex + templates).
- Show issues, suggested fixes, and export a CSV report.
- Optional: apply renames when exporting a ZIP (do not modify originals).
Deliverables:
- Full Next.js TS app, file tree.
- Commands + Docker.
- Tests or QA checklist.
中文释义: 生成一个“图片素材命名规范检查”在线工具：可配置规则、输出报告、导出时应用重命名，并给出源码/部署/QA。
