[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74085":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":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":25,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":29,"readmeContent":30,"aiSummary":31,"trendingCount":16,"starSnapshotCount":16,"syncStatus":32,"lastSyncTime":33,"discoverSource":34},74085,"Jellyfish","Forget-C\u002FJellyfish","Forget-C","An end-to-end production workspace for AI-generated short dramas. From script input to structured storyboarding, consistency management, shot preparation, video generation, and export.","https:\u002F\u002Fforget-c.github.io\u002FJellyfish",null,"Python",3794,693,24,1,0,23,85,533,69,110.52,"Apache License 2.0",false,"main",true,[27,28],"ai","short-drama","2026-06-12 04:01:13","# Jellyfish — AI Short Drama Studio\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\".\u002Fdocs\u002Fimg\u002Flogo.svg\" alt=\"Jellyfish Logo\" width=\"160\" \u002F>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0\">\n    \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache%202.0-blue.svg\" alt=\"License\" \u002F>\n  \u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Ffrontend-React%20%2B%20Vite-61DAFB\">\n    \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Ffrontend-React%20%2B%20Vite-61DAFB\" alt=\"Frontend\" \u002F>\n  \u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fbackend-FastAPI-009688\">\n    \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fbackend-FastAPI-009688\" alt=\"Backend\" \u002F>\n  \u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FForget-C\u002FJellyfish\u002Factions\u002Fworkflows\u002Fdeploy-site.yml\">\n    \u003Cimg src=\"https:\u002F\u002Fgithub.com\u002FForget-C\u002FJellyfish\u002Factions\u002Fworkflows\u002Fdeploy-site.yml\u002Fbadge.svg\" alt=\"Deploy Site\" \u002F>\n  \u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FForget-C\u002FJellyfish\u002Factions\u002Fworkflows\u002Fghcr-images.yml\">\n    \u003Cimg src=\"https:\u002F\u002Fgithub.com\u002FForget-C\u002FJellyfish\u002Factions\u002Fworkflows\u002Fghcr-images.yml\u002Fbadge.svg\" alt=\"Build and push images\" \u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\".\u002FREADME.md\">English\u003C\u002Fa> ·\n  \u003Ca href=\".\u002Fdocs\u002FREADME.ja.md\">日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\nAn end-to-end production workspace for AI-generated short dramas.  \nFrom script input to structured storyboarding, consistency management,\nshot preparation, video generation, and export.\n\n## 📷 Screenshots\n\n| Project overview | Asset management |\n| --- | --- |\n| \u003Cimg src=\".\u002Fdocs\u002Fimg\u002Fproject.png\" alt=\"Project overview\" width=\"420\" \u002F> | \u003Cimg src=\".\u002Fdocs\u002Fimg\u002F%E8%B5%84%E4%BA%A7%E7%AE%A1%E7%90%86.png\" alt=\"Asset management\" width=\"420\" \u002F> |\n\n## ✨ Core Value\n\n- **Connect the full production flow**: Move from script input to storyboard preparation, image\u002Fvideo generation, and task tracking in one place.\n- **Turn AI output into reusable production assets**: Shots, candidate assets, dialogue, prompts, and generation tasks can all be reviewed and reused.\n- **Treat consistency as a first-class problem**: Centralized character, scene, prop, and costume management reduces drift across shots.\n- **Handle long-running generation as trackable tasks**: Text, image, and video jobs all go through one async task system with status, cancel, and recovery.\n- **Build AI capability as infrastructure**: Model management, prompt templates, files, and OpenAPI-based collaboration make the system extensible.\n\n## ✨ Core Capabilities\n\nJellyfish is not just a single “AI image\u002Fvideo” utility. It is a\nproduction workspace built around:\n\n- script understanding\n- shot preparation\n- asset consistency\n- generation execution\n- task tracking\n\n### 1. AI script understanding and storyboard breakdown\n\n- Split chapter scripts into shots\n- Extract characters, scenes, props, costumes, and dialogue\n- Run script optimization, simplification, and consistency checks\n- Support targeted analysis such as character portraits or scene details\n\n### 2. Shot preparation and confirmation workflow\n\nThe main workflow is:\n\n`script breakdown → shot preparation → candidate confirmation → shot ready → generation workspace`\n\nPreparation currently supports:\n\n- extracting and refreshing shot candidates\n- accepting or ignoring asset candidates\n- accepting or ignoring dialogue candidates\n- linking existing characters, scenes, props, and costumes\n- correcting shot-level basic information\n- using a unified readiness state to decide whether a shot is prepared\n\n### 3. Asset consistency and reuse\n\nThe system maintains a shared entity model across:\n\n- characters \u002F actors\n- scenes\n- props\n- costumes\n\nThis supports asset reuse across shots and helps stabilize style and identity.\n\n### 4. Shot-level image and video orchestration\n\nOnce a shot is `ready`, the generation workspace supports:\n\n- keyframe and reference image management\n- shot-level video prompt preview\n- image and video generation tasks\n- single-shot and batch pre-checks\n- writing generation outputs back into the shot\u002Fmedia system\n\n### 5. Unified async task center\n\nCurrent task infrastructure supports:\n\n- async text-processing tasks\n- async image and video generation tasks\n- unified task status, result, and elapsed-time tracking\n- task cancellation\n- a global task center with context-aware navigation back to project\u002Fchapter\u002Fshot\n\n### 6. Model, prompt, and generation infrastructure\n\nSupporting capabilities include:\n\n- multi-provider \u002F multi-model management\n- default model settings by category\n- prompt template management\n- file and generated media management\n- OpenAPI-driven frontend\u002Fbackend contracts\n\n## 🚀 Feature Overview\n\n### Project and chapter management\n\n- Create and manage projects and chapters\n- Use chapters as the unit for scripts, shots, and generation\n- Provide dashboard-style entry points and aggregated stats\n\n### AI script processing\n\n- Break chapter scripts into shots\n- Extract characters, scenes, props, costumes, and dialogue\n- Support optimization, simplification, and consistency checks\n- Support focused analysis such as character portraits or scene information\n\n### Shot preparation workflow\n\n- Edit shot title, summary, and basic information\n- Refresh extracted asset and dialogue candidates\n- Confirm, ignore, or link candidate items\n- Use preparation state to determine shot readiness\n- Keep “prepared” distinct from “currently generating”\n\n### Asset and entity management\n\n- Manage characters, actors, scenes, props, and costumes\n- Link and reuse them at shot level\n- Manage entity images\n- Check name existence to encourage reuse of existing assets\n\n### Shot generation workspace\n\n- Manage keyframes, reference images, and video prompts\n- Check video readiness before generation\n- Launch image\u002Fvideo generation tasks\n- Support both single-shot and batch generation workflows\n\n### Task center\n\n- View active and recently finished tasks\n- Track status, progress, elapsed time, and results\n- Cancel tasks\n- Jump back to the related project, chapter, or shot\n\n### Model and prompt infrastructure\n\n- Manage providers, models, and default settings\n- Manage prompt templates for images, video, and shots\n- Generate frontend request helpers and types from OpenAPI\n- Provide a stable base for future AI workflow expansion\n\n### File and media management\n\n- Manage uploads and generated outputs\n- Preview, link, and reuse image\u002Fvideo assets\n- Preserve shot and entity context around generated media\n\n## 🎯 Use Cases\n\n- Short \u002F micro-drama creators\n- AI studios producing video content in batches\n- Solo creators exploring vertical drama production\n- Education and training teams making lesson videos\n- Brands and e-commerce teams producing story-driven promos\n\n## 🔁 Frontend OpenAPI client and type generation\n\nFrontend request helpers and types are generated from the backend\nOpenAPI spec. Output directory:\n\n- `front\u002Fsrc\u002Fservices\u002Fgenerated\u002F`\n\nCached spec file:\n\n- `front\u002Fopenapi.json`\n\nWith the backend dev server running at `http:\u002F\u002F127.0.0.1:8000`, run:\n\n```bash\ncd front\npnpm run openapi:update\n```\n\n## 🐳 Docker Compose\n\nThe repository includes a ready-to-run compose setup under\n`deploy\u002Fcompose\u002F`.\n\n### Ports\n\n- Frontend: `http:\u002F\u002Flocalhost:7788`\n- Backend: `http:\u002F\u002Flocalhost:8000` (`\u002Fdocs` for Swagger)\n- MySQL: `localhost:${MYSQL_PORT:-3306}`\n- Redis: `localhost:${REDIS_PORT:-6379}`\n- RustFS: `http:\u002F\u002Flocalhost:${RUSTFS_PORT:-9000}`\n\n### Start\n\n```bash\ncp deploy\u002Fcompose\u002F.env.example deploy\u002Fcompose\u002F.env\ndocker compose --env-file deploy\u002Fcompose\u002F.env -f deploy\u002Fcompose\u002Fdocker-compose.yml up --build\n```\n\n## 🧑‍💻 Local Development\n\n### Backend\n\n```bash\ncd backend\ncp .env.example .env\nuv sync\nuv run uvicorn app.main:app --reload --host 0.0.0.0 --port 8000\n```\n\n### Frontend\n\n```bash\ncd front\npnpm install\npnpm dev\n```\n\n## 📄 License\n\nThis project is licensed under [Apache-2.0](.\u002FLICENSE).\n\n## 💬 Community & Feedback\n\n- [GitHub Issues](https:\u002F\u002Fgithub.com\u002FForget-C\u002FJellyfish\u002Fissues)\n\n","Jellyfish 是一个用于生成AI短剧的端到端生产工作空间，涵盖了从剧本输入、分镜头脚本制作、一致性管理、镜头准备到视频生成与导出的全流程。该项目采用Python编写，前端基于React + Vite构建，后端使用FastAPI支持，具备连接整个制作流程的能力，能够将AI生成的内容转化为可复用的生产资源，并通过集中化管理角色、场景等元素确保作品的一致性。此外，它还支持长时间运行的任务追踪和管理。Jellyfish适合于需要高效创作和管理AI驱动的短片或微电影的场景，如独立制片人、内容创作者及小型影视团队。",2,"2026-06-11 03:48:44","high_star"]