About this project

OpenCreator (formerly KrillinAI) is an open-source AI workspace that combines multimodal content creation with a general Agent workspace. It uses Codex CLI as its execution engine and adds a local Runtime, a visual workspace, and a Desktop host around it. The product brings together two connected workflows: 1. AI content creation: dedicated creator tools for video translation, video downloading, thumbnail generation, image generation, smart dubbing, and video generation. 2. General Agent workspace: organize conversations by project, keep Runs working in the background, and manage approvals, attachments, files, Skills, MCP, schedules, notifications, memory, and diagnostics. Creator Tools currently available include: - Video Translation: Import local or public videos, transcribe with cloud or local Whisper, use LLM for subtitle segmentation/alignment/terminology/translation, configure bilingual subtitles, dubbing, subtitle styles, and export SRT/audio/video. - Video Downloader: Parse YouTube, Bilibili, and other public links, inspect quality and format options, download video or audio. - Thumbnail Generator: Combine topic, video link, and optional reference image to generate multiple thumbnail variations. - Image Generation: Generate with GPT Image from prompts and optional reference images, configure aspect ratio and output count. - Smart Dubbing: Turn scripts into voiceovers with selectable voices, pacing, and emotion controls. - Video Generation: Generate videos with Seedance from prompts and reference images. Tools in development include Stick Figure Animation, Auto Clips, and Digital Avatar. The workspace and Agent conversation share a single state machine, keeping steps, progress, and results synchronized. Each revision creates a new version without overwriting earlier results. The system supports global, project, and thread memory with reproducible Run input snapshots. Supported models span language models (GPT, DeepSeek, Qwen, Kimi, GLM, Grok, Doubao, ERNIE, Hunyuan), image (GPT Image), video (Seedance), and voice/transcription (Whisper, OpenAI TTS, MiniMax, Edge TTS, Aliyun Speech). Architecture includes a React 18/Vite/TypeScript frontend, a Fastify-based local daemon Runtime, an Electron Desktop host, and a media toolchain using yt-dlp, FFmpeg, and Whisper. Data is stored locally via SQLite and filesystem, with credentials in system credential storage. The project is licensed under Apache 2.0.