[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-2763":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":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":16,"stars30d":13,"stars90d":15,"forks30d":15,"starsTrendScore":17,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":20,"hasPages":22,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":41,"readmeContent":42,"aiSummary":43,"trendingCount":15,"starSnapshotCount":15,"syncStatus":44,"lastSyncTime":45,"discoverSource":46},2763,"image-prompt-library","EddieTYP\u002Fimage-prompt-library","EddieTYP","Local-first image prompt library for generating images, saving prompts, tags, and collections.","https:\u002F\u002Feddietyp.github.io\u002Fimage-prompt-library\u002F",null,"Python",119,16,11,0,1,3,3.69,"GNU Affero General Public License v3.0",false,"main",true,[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40],"ai-image","all-in-one","allinone","chatgpt","fastapi","gpt-image-2","gpt-image-2-prompts","image-generation","image-prompts","local-first","openai-codex","openai-image-generation","prompt-library","react","self-hosted","sqlite","text-to-image","2026-06-12 02:00:43","# Image Prompt Library\n\n[![CI](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Fworkflows\u002FCI\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Factions\u002Fworkflows\u002Fci.yml)\n[![GitHub Pages demo](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Fworkflows\u002FDeploy%20GitHub%20Pages%20demo\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Factions\u002Fworkflows\u002Fpages.yml)\n[![Release](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Ftag\u002FEddieTYP\u002Fimage-prompt-library?sort=semver&label=release)](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Freleases\u002Ftag\u002Fv0.7.4-beta)\n[![License: AGPL-3.0-or-later](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-AGPL--3.0--or--later-blue)](LICENSE)\n\n\u003Cp align=\"center\">\n  \u003Cstrong>Language:\u003C\u002Fstrong>\n  \u003Cstrong>English\u003C\u002Fstrong> |\n  \u003Ca href=\"README_zh-TW.md\">繁體中文\u003C\u002Fa> |\n  \u003Ca href=\"README_zh-CN.md\">简体中文\u003C\u002Fa>\n\u003C\u002Fp>\n\n**Image Prompt Library** is a local-first visual library for generated images and the prompts behind them. Save useful image results, preserve the prompt and source metadata, organize references into collections and tags, and find them again as an image-first catalogue.\n\nYour private library stays on your machine: local SQLite, local image files, no hosted database, no built-in cloud sync, and no account required.\n\n\n## Introduction\n\nImage Prompt Library is built for the moment when image-generation prompts become reusable knowledge rather than one-off chat messages.\n\n**Browse the read-only online demo:** \u003Chttps:\u002F\u002Feddietyp.github.io\u002Fimage-prompt-library\u002F>\n\nThe public demo is a browsable catalogue of **533 prompt\u002Fimage references** gathered from two generous upstream galleries: [`wuyoscar\u002Fgpt_image_2_skill`](https:\u002F\u002Fgithub.com\u002Fwuyoscar\u002Fgpt_image_2_skill) (**CC BY 4.0**) and [`freestylefly\u002Fawesome-gpt-image-2`](https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fawesome-gpt-image-2) (**MIT**). It covers UI and interface design, posters and typography, product and e-commerce imagery, charts and infographics, technical diagrams, photography, character portraits, architecture, storytelling scenes, and illustration styles. Each reference is shown as an image-first card, with English, Traditional Chinese, and Simplified Chinese prompt variants where available.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Fscreenshots\u002Fpublic-demo-v0.6-533-references.png\" alt=\"Image Prompt Library public demo showing 533 prompt references\" width=\"100%\" \u002F>\n\u003C\u002Fp>\n\nUse the online demo as a visual prompt catalogue: search for ideas, inspect prompt structure, copy public sample prompts, and compare how different prompt styles map to different image outputs. The GitHub Pages demo is intentionally static and read-only: Add\u002FEdit, private library management, and image generation are local-install features.\n\nIf you want to manage your own private prompt\u002Fimage library, install the app locally. Local installs let you add and edit your own images and prompts, organize them into collections and tags, search them later, and optionally generate new images through ChatGPT \u002F Codex OAuth while keeping your SQLite database and image files on your own computer.\n\nCurrent public beta: [`v0.7.4-beta`](https:\u002F\u002Fgithub.com\u002FEddieTYP\u002Fimage-prompt-library\u002Freleases\u002Ftag\u002Fv0.7.4-beta). This release builds on v0.7 prompt variables, Template indicators, generated-result cleanup, search sort operators, queue cancellation, and restart recovery while clarifying queue review state: generated results that are being reused as references are now marked `Used as ref`, including cloned generation-result references, and the queue loads enough recent history to show those relationships consistently.\n\n## Quick start\n\nNormal release installs require **Python 3.10+** and `curl`. They do **not** require Node.js.\n\n```bash\ncurl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002FEddieTYP\u002Fimage-prompt-library\u002Fmain\u002Fscripts\u002Finstall.sh | bash\nimage-prompt-library start\n```\n\n`image-prompt-library start` runs the local server in the current terminal. Keep it open, then visit \u003Chttp:\u002F\u002F127.0.0.1:8000\u002F> in your browser. Press `Ctrl-C` in that terminal to stop the server.\n\nOptional: import a starter sample pack if you want demo references in a fresh local library.\n\n```bash\nimage-prompt-library sample-data en       # English collection names\nimage-prompt-library sample-data zh_hans  # Simplified Chinese collection names\nimage-prompt-library sample-data zh_hant  # Traditional Chinese collection names\n```\n\nThe starter sample pack can be installed with localized collection names in English, Simplified Chinese, or Traditional Chinese. The underlying sample references keep their source titles\u002Fprompts and available prompt variants; this choice mainly affects the imported collection labels and default sample-pack language metadata.\n\nFor the larger Traditional Chinese `awesome-gpt-image-2` sample pack:\n\n```bash\nimage-prompt-library sample-data zh_hant awesome-gpt-image-2\n```\n\nFor update, rollback, service mode, uninstall, WSL, and source-development setup, see [Documentation](#documentation).\n\n## What you can do\n\n- **Browse visually:** scan prompt references in Cards view or Explore view.\n- **Search and filter:** search titles, prompts, tags, collections, sources, and notes; combine search with collection filters.\n- **Preserve prompt provenance:** keep original\u002Fsource prompt variants and translated or converted variants side by side.\n- **Manage a private library:** add\u002Fedit your own prompt cards, result images, optional reference images, tags, notes, source URLs, and collections.\n- **Copy reusable prompts:** open an item, choose the prompt language\u002Fsource variant, and copy it with one click.\n- **Generate locally:** connect optional ChatGPT \u002F Codex OAuth in a local install with a ChatGPT subscription that has image-generation access, generate from new or saved prompts, fill `{{variables}}` before sending template prompts, review results, then attach to the current item or save as a new item.\n- **Stay local-first:** your database and image files remain in your local library directory.\n\n## Searching the library\n\nUse the search box at the top of the app to narrow visible references. In the current release, search is plain keyword search across item titles, prompt text, tags, collection names, source metadata, and notes.\n\nExamples:\n\n```text\napple\nposter design\nproduct photo\nawesome-gpt-image-2\n電商\n```\n\nSearch also works with collection filters: choose a collection from **Filters**, then type a keyword to search inside that collection.\n\n## Local generation\n\nLocal installs can optionally connect ChatGPT \u002F Codex OAuth and generate images without adding an OpenAI API key to the app. You will need a ChatGPT account\u002Fsubscription with access to image generation.\n\nBasic flow:\n\n1. Start the local app and open **Config**.\n2. Connect **ChatGPT \u002F Codex OAuth** and approve the device-login flow in your browser.\n3. Return to Image Prompt Library and generate from a new prompt or from an existing saved reference. Prompts can include variables such as `{{subject}}` or `{{style}}`; the composer asks for values before sending the final prompt.\n4. Review generated results in the local inbox.\n5. Attach the result to the current item, or save it as a new item with editable metadata.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Fscreenshots\u002Fgeneration-provider-connected.png\" alt=\"Config drawer showing ChatGPT \u002F Codex OAuth connected for local image generation\" width=\"360\" \u002F>\n\u003C\u002Fp>\n\nThe public GitHub Pages demo never performs live generation and does not expose mutation controls.\n\nFor current generation behavior, limitations, and benchmark notes, see [`docs\u002FGENERATION.md`](docs\u002FGENERATION.md).\n\n## Sample data and attribution\n\nFor first-time setup, Image Prompt Library can import optional sample bundles so you have real prompt\u002Fimage references to explore right away. These samples come from upstream open projects and are included with clear links, thanks, and license notes. They are not presented as original Image Prompt Library artwork or prompts; they remain connected to their original creators and licenses.\n\n| Sample source | License | Notes |\n| --- | --- | --- |\n| [`wuyoscar\u002Fgpt_image_2_skill`](https:\u002F\u002Fgithub.com\u002Fwuyoscar\u002Fgpt_image_2_skill) | CC BY 4.0 | First public sample package and default starter sample library. |\n| [`freestylefly\u002Fawesome-gpt-image-2`](https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fawesome-gpt-image-2) | MIT | Larger Chinese prompt\u002Fimage gallery used by the current public demo and optional sample pack. |\n\nThank you to both upstream projects for making these galleries available. Their prompts and images keep their own source links, attribution, and license terms. Image Prompt Library only provides the local app, import workflow, and browsing\u002Fmanagement interface around them; the app code remains licensed separately under AGPL-3.0-or-later.\n\nFor sample package details and checksums, see [`sample-data\u002FREADME.md`](sample-data\u002FREADME.md).\n\n## Documentation\n\n- [`docs\u002FINSTALLATION.md`](docs\u002FINSTALLATION.md) — install, update, rollback, service mode, uninstall, platform notes.\n- [`docs\u002FGENERATION.md`](docs\u002FGENERATION.md) — ChatGPT \u002F Codex OAuth generation workflow, result review, current limitations, benchmark link.\n- [`docs\u002FDEVELOPMENT.md`](docs\u002FDEVELOPMENT.md) — source setup, dev mode, configuration, data layout, backups.\n- [`docs\u002FTROUBLESHOOTING.md`](docs\u002FTROUBLESHOOTING.md) — common runtime and setup issues.\n- [`CONTRIBUTING.md`](CONTRIBUTING.md) — contributor setup, tests, and project structure.\n- [`ROADMAP.md`](ROADMAP.md) — planned work and project direction.\n\n## License, privacy, and allowed use\n\nImage Prompt Library's core application code is open source under **AGPL-3.0-or-later**. Copyright (C) 2026 Edward Tsoi. See [`NOTICE`](NOTICE) and [`LICENSE`](LICENSE).\n\nCommercial licenses are available for organizations that want to use, modify, or host Image Prompt Library under terms outside the AGPL. Contact the maintainer if you need proprietary hosted-product terms or other non-AGPL licensing.\n\nPrivacy model:\n\n- The app is local-first and stores data on your device.\n- There are no hosted user accounts or built-in cloud sync.\n- Binding to `127.0.0.1` keeps the app local to your machine. Only change the host if you understand LAN exposure.\n\n## Project status\n\nThis is a public beta. Core browsing, search, local add\u002Fedit, optional local generation, versioned installs, update\u002Frollback, and the read-only online demo are available today. Remaining work includes service\u002Fupdate hardening, management-mode cleanup tools, search\u002Fsort polish, batch image editing, import-flow polish, and deeper mobile Explore gestures.\n","Image Prompt Library 是一个本地优先的图像提示库，用于生成图像、保存提示、标签和集合。该项目使用 Python 开发，并结合了 FastAPI 和 React 技术栈来提供后端服务与前端界面，数据存储基于 SQLite 数据库。用户可以保存有价值的图像结果及其背后的提示信息，组织成集合或打上标签以便于管理和查找。适合需要在本地环境中构建个性化图像生成资源库的场景，如设计师、艺术家或是任何对AI图像生成感兴趣的人士。无需云同步或账号注册，所有内容均保存于本地机器上，确保了数据的安全性和私密性。",2,"2026-06-11 02:51:10","CREATED_QUERY"]