Summary

This repository provides an AI-powered command-line video editor that automates the creation of polished advertisements from raw footage and a creative brief. It targets users needing efficient video production, leveraging an ensemble of AI agents and FFmpeg to handle scene analysis, shot selection, assembly, and quality review.

How It Works

The system ingests raw footage and a creative brief, then preprocesses clips for scene detection, transcription, and shot indexing. An AI 'Director' agent, guided by style templates, selects and orders shots, generating an 'EditPlan'. A 'Trim Refiner' optimizes shot boundaries, and an 'Editor' renders the video using FFmpeg. A 'Reviewer' agent scores the output against criteria like adherence and watchability, feeding back to the Director for iterative improvements if quality thresholds are not met.

Quick Start & Requirements

• Install: Clone the repository, create and activate a Python 3.11+ virtual environment, and install dependencies using pip install -e ".[dev]".
• Prerequisites: Python 3.11+, FFmpeg installed and accessible in the system's PATH, and a Google AI API key.
• Usage: Execute edits via the ave edit CLI command, specifying footage directory, creative brief (JSON or file path), pipeline YAML, and style YAML.
• Docs: Primary documentation is within the README; links to official quick-start or demo pages are not explicitly provided.

Highlighted Details

• An ensemble of AI agents (Director, Trim Refiner, Editor, Reviewer) powered by Google Gemini orchestrates the video editing process.
• Configurable pipelines defined in YAML allow customization of agent steps, retry conditions, and feedback loops.
• Style templates (YAML) provide structured guidance for creative elements like segment durations, pacing, and text overlays.
• Automated quality review with iterative refinement based on multiple scoring dimensions (Adherence, Pacing, Visual Quality, Watchability, Overall).
• The command-line interface (CLI) is the primary, supported interaction method.

Maintenance & Community

The README does not detail specific contributors, sponsorships, or community channels (e.g., Discord, Slack).

Licensing & Compatibility

The project is released under the MIT License, which is generally permissive for commercial use and integration into closed-source projects.

Limitations & Caveats

The experimental Web UI (AVE Studio) is explicitly marked as pre-alpha, a work in progress, and not recommended for production use, with expected instability and missing features. The CLI remains the stable, supported interface.