Buzz AI: The Complete Human-and-Agent Workspace Course

This course contains the use of artificial intelligence.

Buzz brings people, AI agents, shared knowledge, repositories, and automation into one workspace. This course gives you a complete, practical path from understanding the platform to operating a secure human-agent system with clear ownership and verification.

You will begin with the core Buzz model: communities, relays, channels, direct messages, Pulse, canvases, projects, agents, and workflows. Then you will follow the official download and onboarding path, protect your identity, organize a clean workspace, and learn how messages, threads, files, reminders, notifications, and shared notes behave in real use.

The agent modules show how to connect and operate Codex and Hermes, evaluate other runtimes, define focused personas, restrict permissions, choose working directories, manage memory, and prevent accidental agent-to-agent loops. You will see the difference between posting an ordinary message and invoking a real agent identity, with visible replies and explicit recovery when a run fails.

From there, the course moves into repositories and automation. You will coordinate issues, patches, pull requests, reviews, triggers, actions, webhooks, schedules, YAML definitions, approval gates, run history, failure recovery, and audit evidence. The final modules cover the Buzz CLI, self-hosted relay responsibilities, security, privacy, backups, moderation, and a capstone human-agent operating room.

Every lesson uses real screen operation and approved English narration. Downloadable checklists, prompts, templates, and workflow examples are included with their matching lessons. Consequential actions remain behind a visible human decision instead of being treated as automatically approved.

The course is designed for complete beginners as well as developers, technical leads, product teams, researchers, operations professionals, and advanced agent users. No previous Buzz experience is required. A Windows computer is recommended for following the demonstrations, while official macOS and Linux paths are explained where relevant.