Completed
Introduction: Agenda and the "Agent" definition
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Claude Agent SDK Full Workshop - AI-Powered Development Workflows
Automatically move to the next video in the Classroom when playback concludes
- 1 Introduction: Agenda and the "Agent" definition
- 2 The "Harness" concept: Tools, Prompts, and Skills
- 3 Live Coding Setup: Initializing the Agent class and environment
- 4 implementing the "Think" step: Getting the model to reason before acting
- 5 The Agent Loop: connecting `act`, `observe`, and `loop`
- 6 Tool Execution: Handling XML parsing and tool inputs
- 7 The "Bash" Tool: Giving the agent command line access
- 8 Safety & Permissions: "ReadOnly" vs "ReadWrite" file access
- 9 Context Engineering: Using `ls` and `cat` to build dynamic context
- 10 The "Monitor": Viewing the agent's thought process in real-time
- 11 Handling "Stuck" States: Feedback loops and error correction
- 12 Multi-turn Complex Tasks: Building a "Research Agent" demo
- 13 Refactoring patterns: "Hooks" and deterministic overrides
- 14 Q&A: Reproducibility, helper scripts, and non-determinism
- 15 Q&A: Strategies for massive codebases 50M+ lines
- 16 Closing remarks and future SDK roadmap
