{"name":"Create a high-quality native mobile app, \"Readdit,\" that transforms Reddit threads into AI-generated audio or video podcasts. The app must have a polished, professional design and a robust feature set.\n\nThe app will consist of four primary sections:\n\n1.  **Home Screen:**\n    *   A prominent text input field for pasting a Reddit post URL.\n    *   A \"Generate\" button that, when tapped, proceeds to a \"Generation Options\" screen.\n\n2.  **Generation Options Screen:**\n    *   After a URL is entered, this screen allows the user to customize the output.\n    *   Provide a choice between \"Audio Podcast\" and \"Video Podcast.\"\n    *   Include a slider or dropdown to select the number of top comments to process (e.g., 10, 25, 50).\n    *   A \"Start Generation\" button. Tapping this adds the job to a queue, navigates the user to the Library screen, and begins generation in the background.\n\n3.  **Library Screen:**\n    *   The main hub of the app, displaying a list of the user's podcasts.\n    *   Include a search bar at the top to filter podcasts by title.\n    *   Each list item should display the podcast title, creation date, and its status (e.g., \"In Progress,\" \"Complete\"). For \"In Progress\" items, show a visual progress indicator.\n    *   Users should be able to swipe on an item to reveal a \"Delete\" option.\n\n4.  **Media Player Screen:**\n    *   Opens when a completed podcast is tapped in the library.\n    *   For **Video**: Display simple, animated avatars for each AI speaker. The currently speaking character's avatar should be visually highlighted (e.g., a glowing border or subtle animation).\n    *   For **Audio**: Display the post title and the \"username\" of the AI persona currently speaking.\n    *   **Standard Controls**: Include a play/pause button, a seek bar, and playback speed controls (1x, 1.5x, 2x).\n    *   **Context & Sharing**: Must include a \"View Original Post\" button to open the source Reddit URL and a \"Share\" button to generate a shareable link.\n\nBackend and AI Requirements:\n*   The backend must support a job queue for handling concurrent podcast generations.\n*   The generation process should be a background task, and the app must use push notifications to alert the user upon completion.\n*   **AI Persona Engine**: For each thread, create distinct AI personas for the narrator, original poster, and commenters. Each persona needs a unique and highly realistic human voice, personality, and natural speech patterns.\n*   **Video Generation**: If selected, generate a simple video file with animated avatars corresponding to the audio track.\n*   All user-generated content (podcasts, settings) must be stored persistently in a database and linked to the active user's account.\n","short_name":"Create a high-quality native mobile app, \"Readdit,\" that transforms Reddit threads into AI-generated audio or video podcasts. The app must have a polished, professional design and a robust feature set.\n\nThe app will consist of four primary sections:\n\n1.  **Home Screen:**\n    *   A prominent text input field for pasting a Reddit post URL.\n    *   A \"Generate\" button that, when tapped, proceeds to a \"Generation Options\" screen.\n\n2.  **Generation Options Screen:**\n    *   After a URL is entered, this screen allows the user to customize the output.\n    *   Provide a choice between \"Audio Podcast\" and \"Video Podcast.\"\n    *   Include a slider or dropdown to select the number of top comments to process (e.g., 10, 25, 50).\n    *   A \"Start Generation\" button. Tapping this adds the job to a queue, navigates the user to the Library screen, and begins generation in the background.\n\n3.  **Library Screen:**\n    *   The main hub of the app, displaying a list of the user's podcasts.\n    *   Include a search bar at the top to filter podcasts by title.\n    *   Each list item should display the podcast title, creation date, and its status (e.g., \"In Progress,\" \"Complete\"). For \"In Progress\" items, show a visual progress indicator.\n    *   Users should be able to swipe on an item to reveal a \"Delete\" option.\n\n4.  **Media Player Screen:**\n    *   Opens when a completed podcast is tapped in the library.\n    *   For **Video**: Display simple, animated avatars for each AI speaker. The currently speaking character's avatar should be visually highlighted (e.g., a glowing border or subtle animation).\n    *   For **Audio**: Display the post title and the \"username\" of the AI persona currently speaking.\n    *   **Standard Controls**: Include a play/pause button, a seek bar, and playback speed controls (1x, 1.5x, 2x).\n    *   **Context & Sharing**: Must include a \"View Original Post\" button to open the source Reddit URL and a \"Share\" button to generate a shareable link.\n\nBackend and AI Requirements:\n*   The backend must support a job queue for handling concurrent podcast generations.\n*   The generation process should be a background task, and the app must use push notifications to alert the user upon completion.\n*   **AI Persona Engine**: For each thread, create distinct AI personas for the narrator, original poster, and commenters. Each persona needs a unique and highly realistic human voice, personality, and natural speech patterns.\n*   **Video Generation**: If selected, generate a simple video file with animated avatars corresponding to the audio track.\n*   All user-generated content (podcasts, settings) must be stored persistently in a database and linked to the active user's account.\n","description":"Create a high-quality native mobile app, \"Readdit,\" that transforms Reddit threads into AI-generated audio or video podcasts. The app must have a polished, professional design and a robust feature set.\n\nThe app will consist of four primary sections:\n\n1.  **Home Screen:**\n    *   A prominent text input field for pasting a Reddit post URL.\n    *   A \"Generate\" button that, when tapped, proceeds to a \"Generation Options\" screen.\n\n2.  **Generation Options Screen:**\n    *   After a URL is entered, this screen allows the user to customize the output.\n    *   Provide a choice between \"Audio Podcast\" and \"Video Podcast.\"\n    *   Include a slider or dropdown to select the number of top comments to process (e.g., 10, 25, 50).\n    *   A \"Start Generation\" button. Tapping this adds the job to a queue, navigates the user to the Library screen, and begins generation in the background.\n\n3.  **Library Screen:**\n    *   The main hub of the app, displaying a list of the user's podcasts.\n    *   Include a search bar at the top to filter podcasts by title.\n    *   Each list item should display the podcast title, creation date, and its status (e.g., \"In Progress,\" \"Complete\"). For \"In Progress\" items, show a visual progress indicator.\n    *   Users should be able to swipe on an item to reveal a \"Delete\" option.\n\n4.  **Media Player Screen:**\n    *   Opens when a completed podcast is tapped in the library.\n    *   For **Video**: Display simple, animated avatars for each AI speaker. The currently speaking character's avatar should be visually highlighted (e.g., a glowing border or subtle animation).\n    *   For **Audio**: Display the post title and the \"username\" of the AI persona currently speaking.\n    *   **Standard Controls**: Include a play/pause button, a seek bar, and playback speed controls (1x, 1.5x, 2x).\n    *   **Context & Sharing**: Must include a \"View Original Post\" button to open the source Reddit URL and a \"Share\" button to generate a shareable link.\n\nBackend and AI Requirements:\n*   The backend must support a job queue for handling concurrent podcast generations.\n*   The generation process should be a background task, and the app must use push notifications to alert the user upon completion.\n*   **AI Persona Engine**: For each thread, create distinct AI personas for the narrator, original poster, and commenters. Each persona needs a unique and highly realistic human voice, personality, and natural speech patterns.\n*   **Video Generation**: If selected, generate a simple video file with animated avatars corresponding to the audio track.\n*   All user-generated content (podcasts, settings) must be stored persistently in a database and linked to the active user's account.\n","start_url":"/s/284691?addToHomeScreen","display":"standalone","background_color":"#ffffff","theme_color":"#ffffff","icons":[{"src":"https://s.asim.sh/icon-pwa.png","sizes":"192x192","type":"image/png"},{"src":"https://s.asim.sh/icon-pwa.png","sizes":"512x512","type":"image/png"}]}