BestHub
Sign in
Tagged articles

AI Agent Architecture

10 articles · Page 1 of 1
Architecture Development Notes
Architecture Development Notes
Sep 7, 2026 · Artificial Intelligence

From Chat to Control Plane: Making Agent Sessions Addressable and Orchestratable

This article explores how coding agent sessions are evolving into addressable runtime objects with a control plane enabling external message injection, session overview, inter-session references, and persistent metadata — detailing the architectural shift from chat-bound terminals to orchestratable services, plus engineering challenges like metadata persistence, directory boundaries as security perimeters, external write authentication, and queryable session states.

AI Agent ArchitectureAgent SessionsControl Plane
0 likes · 15 min read
From Chat to Control Plane: Making Agent Sessions Addressable and Orchestratable
Architecture Development Notes
Architecture Development Notes
Sep 4, 2026 · Artificial Intelligence

75% Cheaper Cache Reads: Why Long-Running Agent Costs Now Depend on Prefix Stability

Anthropic's Fable 5.1 reduces cache read pricing from $1 to $0.25 per million tokens, shifting long-running agent cost bottlenecks from output to repeated stable prefix reads, making prefix stability, cache breakpoint placement, TTL tuning, and hit-rate observability critical architectural levers for cost control.

AI Agent ArchitectureAnthropicContext Management
0 likes · 15 min read
75% Cheaper Cache Reads: Why Long-Running Agent Costs Now Depend on Prefix Stability
Linyb Geek Road
Linyb Geek Road
Jul 14, 2026 · Artificial Intelligence

Understanding MCP, Skill, Harness, and Loop: A Deep Dive into the Four‑Layer AI Agent Architecture

The article breaks down the four‑layer AI Agent stack—MCP protocol, Agent Skill, Harness runtime, and Loop engineering—showing how each layer solves distinct problems, presenting benchmark data (e.g., a 25.7 pp SWE‑bench gain from Harness changes), security analyses, design trade‑offs, and a production checklist.

AI Agent ArchitectureAgent HarnessAgent Skill
0 likes · 26 min read
Understanding MCP, Skill, Harness, and Loop: A Deep Dive into the Four‑Layer AI Agent Architecture
LuTiao Programming
LuTiao Programming
Jun 17, 2026 · Backend Development

Why Salesforce’s $3.6B AI Customer Service Bet Highlights the Real Opportunity for Java Back‑End Developers

The article explains how Salesforce’s $3.6 billion acquisition of Fin signals a shift from simple chatbot answers to AI agents that execute end‑to‑end business actions, and why Java/Spring Boot developers must expose secure, auditable services rather than merely wrapping large‑model APIs.

AI Agent ArchitectureAI Customer ServiceJava
0 likes · 21 min read
Why Salesforce’s $3.6B AI Customer Service Bet Highlights the Real Opportunity for Java Back‑End Developers
James' Growth Diary
James' Growth Diary
May 21, 2026 · Artificial Intelligence

Why Hermes Agent Stands Out: From One‑Shot Tool to Long‑Term Partner

The article explains how Hermes Agent redesigns AI agents to grow like a partner—adding persistent multi‑layer memory, autonomous skill learning, model‑agnostic architecture, multi‑platform unification and safe autonomous behavior—addressing the shortcomings of typical one‑off AI tools.

AI Agent ArchitectureHermes AgentModel-agnostic
0 likes · 12 min read
Why Hermes Agent Stands Out: From One‑Shot Tool to Long‑Term Partner
Alibaba Cloud Developer
Alibaba Cloud Developer
Apr 23, 2026 · Artificial Intelligence

How Hermes Agent Achieves Self‑Improving AI Through Memory, Skills, and Nudge Engine

Hermes Agent combines a bounded memory store, automatically generated reusable skills, and a Nudge Engine that periodically triggers background reviews, forming a self‑improving loop that reduces tool calls, fixes recurring errors, and outperforms OpenClaw’s static skill system.

AI Agent ArchitectureHermes AgentMemory Management
0 likes · 28 min read
How Hermes Agent Achieves Self‑Improving AI Through Memory, Skills, and Nudge Engine
Cognitive Technology Team
Cognitive Technology Team
Apr 15, 2026 · Artificial Intelligence

Why Claude Code’s Architecture Keeps Agents Stable: A Deep Dive into Runtime Design

This article dissects Claude Code’s multi‑layered architecture—entry routing, REPL orchestration, query loop, tool runtime, permission system, task management, and extension layers—to reveal how each layer isolates complexity, enabling robust, long‑running AI agents that scale without collapsing under real‑world workloads.

AI Agent ArchitectureClaude CodePermission System
0 likes · 42 min read
Why Claude Code’s Architecture Keeps Agents Stable: A Deep Dive into Runtime Design
AI Explorer
AI Explorer
Apr 1, 2026 · Artificial Intelligence

Inside the Claude Code Source Leak: 1,900 Files, 510k Lines, and the Three‑Layer Memory Architecture

The March 31, 2026 accidental npm leak of Claude Code's source revealed over 1,900 TypeScript files and a three‑layer memory design, exposed unreleased features, disclosed a concurrent axios supply‑chain attack, and prompted concrete security and engineering lessons for AI developers.

AI Agent ArchitectureClaude Codenpm supply chain
0 likes · 8 min read
Inside the Claude Code Source Leak: 1,900 Files, 510k Lines, and the Three‑Layer Memory Architecture
Data Party THU
Data Party THU
Mar 27, 2026 · Artificial Intelligence

When AI Steps Out of the Screen: How Looki’s Proactive Engine Turns Wearables into Real‑World Assistants

The article analyzes the limits of screen‑bound AI agents, explains how Looki’s Proactive Intelligence Engine extends OpenClaw’s capabilities into the physical world through context‑aware perception, dynamic decision‑making, and privacy‑preserving pipelines, and discusses the technical and practical challenges of building truly proactive AI.

AI Agent ArchitectureLookiOpenClaw
0 likes · 11 min read
When AI Steps Out of the Screen: How Looki’s Proactive Engine Turns Wearables into Real‑World Assistants