GoAgent Source Deep Dive Series

This directory contains the English drafts for the GoAgent blog series. All diagrams use Mermaid for blog-friendly publishing.

Narrative Thread

Every article follows the same thread: Problem → Limitations of existing approaches → GoAgent's approach → Architecture naturally emerges.

1. Problem: Start with a concrete pain point — readers immediately understand why the module exists.
2. Limitations: List 1-2 common approaches and explain why they fall short.
3. GoAgent's approach: Present the solution as a natural response to the problem, not a pre-designed architecture.
4. Architecture naturally emerges: From the approach, the architecture follows as a natural derivation.

Reading Order

1. API Client: Converging Framework Capabilities Into a Usable Entry Point — How should internal modules be called externally
2. LLM Client: Unifying Multi-Model Call Boundaries — How to keep agents from being tied to one provider
3. Agent System: The Leader/Sub-Agent Collaboration Skeleton — One agent can't handle everything
4. AHP Protocol: Messages, Queues, and Heartbeats — How do agents communicate
5. Memory System: Sessions, Tasks, and Distilled Memories — Agents start from scratch every conversation
6. Tool System: Registration, Capability Matching, and Execution — How agents "do things" not just "talk"
7. Workflow Engine: DAG-Based Agent Orchestration — How do multi-step tasks execute by dependency
8. Storage and Retrieval: PostgreSQL, pgvector, and Hybrid Search — Where do agent memories and knowledge live
9. Embedding Service: The Engineering Boundary of Vector Generation — How does text become vectors

Writing Principles

• Describe implemented source-code behavior, not wishful roadmap items.
• Include code references so readers can verify claims in the repository.
• Use Mermaid diagrams only; no screenshots or image dependencies.
• Analyze design trade-offs, not just advantages.
• Style reference: In-depth LLVM IR.