学习路径
第 0 课到第 13 课：使用 OpenAI Java SDK 渐进式 Agent 设计
层次图例
Basic Chat LoopThe Foundation of Every Agent
“The simplest agent is just a while loop calling the model”
了解更多Tool UseFunction Calling Basics
“Function calling turns a chat model into an agent that can act”
了解更多ToolsOne Handler Per Tool
“The loop stays the same; new tools register into the dispatch map”
了解更多TodoWritePlan Before You Act
“An agent without a plan drifts; list the steps first, then execute”
了解更多SubagentsClean Context Per Subtask
“Subagents use independent context, keeping the main conversation clean”
了解更多SkillsLoad on Demand
“Inject knowledge via tool_result when needed, not upfront in the system prompt”
了解更多CompactThree-Layer Compression
“Context will fill up; three-layer compression enables infinite sessions”
了解更多TasksTask Graph + Dependencies
“A file-based task graph with ordering, parallelism, and dependencies”
了解更多Background TasksBackground Threads + Notifications
“Run slow operations in the background; the agent keeps thinking ahead”
了解更多Agent TeamsTeammates + Mailboxes
“When one agent can't finish, delegate to persistent teammates via async mailboxes”
了解更多Team ProtocolsShared Communication Rules
“One request-response pattern drives all team negotiation”
了解更多Autonomous AgentsScan Board, Claim Tasks
“Teammates scan the board and claim tasks themselves; no need for assignment”
了解更多Worktree IsolationIsolate by Directory
“Each works in its own directory; tasks manage goals, worktrees manage directories”
了解更多Full Reference AgentCapstone — Everything Combined
“The capstone combines every mechanism from Lessons 1–12 into one agent”
了解更多