Deep dive into agent memory mechanics: five memory files, four context assembly strategies, TOOLS.md as a dual-source responsibility, and Dreaming — background consolidation from Anthropic (Auto Dream, Dreams API, KAIROS) and OpenAI (Dreaming V3, 82.8% factual recall). Engineering decisions behind a production platform.
Five multi-agent orchestration architectures — Supervisor, Swarm, Debate, Assembly Line, Hierarchical — with Mermaid diagrams, code examples in Eino (Go), and an honest look at limitations. When one agent isn’t enough, and when you’re better off staying solo.
Context window ≠ memory. I break down agent amnesia, the academic taxonomy (Du et al., 2026), five approaches to memory — from OpenClaw’s file-based brain to mem0’s managed layer — and how our team solves this in practice.
Plan-and-Execute separates strategic planning from tactical execution. I break down the Planner+Executor+Reviser architecture, ReWOO and LLMCompiler variants with numbers from original papers, prompt injection defense via control-flow integrity, and a minimal working Go example with Eino.
Reflexion is a pattern where an agent learns from its own mistakes through verbal reinforcement and episodic memory. I trace the evolution of self-critique from Constitutional AI to Reflexion, honestly show its limitations (doesn’t work without an external verifier), and implement a working example in Go using Eino compose.Graph with an Actor → Evaluator → Reflector → retry loop.