youtube-automation-agent
🎬 Open-source, self-hosted AI agent that runs a YouTube channel end to end: researches topics, writes scripts, generates narration, visuals and video, optimize
What is youtube-automation-agent?
🎬 Open-source, self-hosted AI agent that runs a YouTube channel end to end: researches topics, writes scripts, generates narration, visuals and video, optimizes SEO, and publishes after your approval, then learns from analytics. Works with Gemini, OpenAI, OpenRouter, DeepSeek and more.
Overview of youtube-automation-agent
What it is
AgentTube is an open-source AI agent that automates end-to-end YouTube channel management. It handles topic research, script writing, narration generation, visual creation, video assembly, metadata optimization, review, scheduling, publishing, and analytics learning.
Key features
- Self-hosted: credentials, media, and channel data stay under your control
- Approval-first: nothing is scheduled until quality, rights, and human-review gates pass by default
- Strategy-driven: set objectives, audience, pillars, cadence, and guardrails for the Autonomous Channel Operator
- Provider-flexible: supports Gemini, OpenAI, OpenRouter, Kimi, MiMo, GLM, and other OpenAI-compatible endpoints; plus Seedance, MiniMax H3, Gemini Omni Flash, Kling, Wan, or local FFmpeg for video
- Observable: dashboard tracks generation jobs, failures, review state, publishing, and local activation milestones
- Works with a free Gemini API or OpenAI
Use cases
Running a fully automated YouTube channel with minimal human intervention. Creating, optimizing, and publishing videos continuously. Managing multiple production stages from research to publishing with approval workflows. Applying discoverability audits and controlled growth experiments.
Getting started
bashgit clone https://github.com/darkzOGx/youtube-automation-agent.gitcd youtube-automation-agentnpm installnpm run walkthroughnpm startOpen http://localhost:3456. The walkthrough explains provider choices, tests credentials, and guides YouTube authorization. An alternative shorter setup is npm run setup, and .env.example documents every setting.
License and pricing
License: MIT. Pricing type: open source.
Based on the project's public README and repository metadata.
