Skip to content

AutoShorts is a local-first desktop application for turning long-form video or audio recordings into high-impact, vertical short-form clip candidates (9:16 portrait) with AI-powered viral moment ranking.

Rust

docs: rebrand Shortcast to AutoShorts and update installation and configuration instructions

JayWebtech291094ceddaa
diff --git a/README.md b/README.md
index a523a6f..352d175 100644
--- a/README.md
+++ b/README.md
@@ -1,33 +1,90 @@
-# Shortcast
+# AutoShorts
-Shortcast is a local-first desktop app for turning long-form recordings into ranked short-form clip candidates, then rendering vertical clips with captions and platform copy.
+AutoShorts is a local-first desktop application for turning long-form video or audio recordings into high-impact, vertical short-form clip candidates (9:16 portrait) with AI-powered viral moment ranking.
-This repository implements the build specification from `Shortcast_Build_Specification.docx` as a Tauri 2 + React + Rust app foundation.
+This repository implements the desktop app foundation using **Tauri 2 + React + TSX + Rust + SQLite**.
-## Current Scope
+---
-- Desktop shell and project UI
-- SQLite data model for projects, transcripts, candidates, clips, copy, and schedule entries
-- Media probing hooks through `ffprobe`
-- Audio extraction/rendering hooks through `ffmpeg`
-- Cloud transcription integration boundary for Deepgram
-- Claude moment-detection integration boundary with structured candidate storage
-- Debug transcript and candidate views
-- Clip-count slider that promotes already-ranked candidates without another LLM call
+## Key Features
+
+- **Dynamic Multi-LLM Support**: Supports both **DeepSeek** (default) and **Claude** (anthropic) for viral moment detection and hooks analysis.
+- **Automated Pipeline**: Imports media, extracts audio, transcribes using Deepgram, and automatically analyzes and ranks moments in a single automated chain.
+- **Local SQLite Storage**: Saves transcripts, candidates, custom names, and rendering data locally.
+- **Native Project Manager**: Create, open, rename, and delete projects from the dashboard.
+- **Portrait Auto-Cropping**: Automatically center-crops landscape videos to vertical H.264 portrait clips using native `ffmpeg` integration.
+- **Key Warnings**: Built-in visual warnings that identify missing environment variables and prompt you directly in the UI.
+
+---
## Prerequisites
-- Node.js 20+
-- Rust 1.80+
-- FFmpeg and FFprobe on `PATH`
-- Optional: `DEEPGRAM_API_KEY`
-- Optional: `ANTHROPIC_API_KEY`
+To run or build the application from source:
+
+- **Node.js**: `20+`
+- **Rust**: `1.80+`
+- **FFmpeg & FFprobe**: Must be installed and available on your system `PATH`.
+ - On macOS, you can install them using Homebrew:
+ ```bash
+ brew install ffmpeg
+ ```
+
+---
+
+## Installation Guide (For Users)
+
+### 1. Download the App Bundle
+Download the latest `.dmg` or `.app` release from the [GitHub Releases](https://github.com/JayWebtech/autoshorts/releases/tag/autoshorts). Be sure to select the architecture matching your Mac:
+- **Apple Silicon (M1/M2/M3)**: Choose the `aarch64` package.
+- **Intel Mac**: Choose the `x64` package.
+
+### 2. Install the App
+1. Double-click the downloaded `.dmg` file.
+2. Drag the **AutoShorts** app icon into your **Applications** folder.
+
+### 3. Bypass macOS Gatekeeper (Unsigned Local Builds)
+Because the local build is self-signed/ad-hoc signed, macOS may block it on first run with a warning ("damaged or from an unidentified developer").
+- **Easiest fix**: Right-click the `AutoShorts.app` icon in Finder, choose **Open**, and click **Open** in the prompt.
+- **Terminal fix**: Run the following command in your terminal:
+ ```bash
+ xattr -cr /Applications/AutoShorts.app
+ ```
+
+### 4. Configure API Credentials
+Since the standalone app runs outside your local environment, you need to configure your API keys within the app interface:
+1. Open **AutoShorts** and click **API Settings** in the top bar.
+2. Paste your API keys for **Deepgram**, **Claude**, and/or **DeepSeek**.
+3. The app will visually badge your active provider and alert you if keys are missing.
-## Run
+---
+## Developer Guide
+
+### 1. Setup Environment Configuration
+Copy `.env.example` to `.env` in the root folder:
+```bash
+cp .env.example .env
+```
+Fill in your API Keys:
+```env
+DEEPGRAM_API_KEY=your-deepgram-api-key
+DEEPSEEK_API_KEY=your-deepseek-api-key
+ANTHROPIC_API_KEY=your-anthropic-api-key
+
+# Choose your default AI analysis provider ("deepseek" or "claude")
+LLM_PROVIDER=deepseek
+```
+
+### 2. Run in Development Mode
+To start the live-reloaded frontend and backend development shell:
```bash
npm install
npm run tauri:dev
```
-If FFmpeg or cloud API keys are unavailable, the UI shows the missing capability and keeps local project/candidate management usable.
+### 3. Build the Application
+To build and package the native macOS app bundle (`.app` and `.dmg` installer):
+```bash
+npm run tauri:build
+```
+The output installers will be built under `src-tauri/target/release/bundle/`.