OverviewOpenClaw reportedly contains around 400,000 lines of code, making direct reading and comprehension quite challenging. Therefore, this series uses Nanobot to learn OpenClaw's distinctive features.Nanobot is an ultra-lightweight personal AI assistant framework open-sourced by the HKU Data Science Laboratory (HKUDS), positioned as an "Ultra-Lightweight OpenClaw." It's perfectly suited for learning Agent architecture.Rich contextual information forms the foundation for effective Agent planning and action. An Agent requires access to vari...
2026年4月
Technical Information PublicationContextBuilder Architecture: The Central Hub for AI Agent Context Management in Nanobot Framework
Executive SummaryOpenClaw comprises approximately 400,000 lines of code, making comprehensive reading and comprehension exceedingly challenging. Therefore, this series explores OpenClaw's distinctive features through Nanobot, an ultra-lightweight personal AI assistant framework open-sourced by Hong Kong University Data Science Laboratory (HKUDS), positioned as "Ultra-Lightweight OpenClaw"—ideal for learning Agent architecture.Rich contextual information forms the foundation for effective Agent planning and action. An Agent requires access to...
Vision Transformer: Bridging Sequence Modeling and Visual Understanding Through Pure Attention Mechanisms
Introduction: From NLP Breakthrough to Visual RevolutionIn our previous exploration, we thoroughly examined the original Transformer architecture and its overall propagation logic. The results speak for themselves: Transformer brought paradigm-shifting breakthroughs to the NLP field by achieving global sequence modeling capabilities through self-attention mechanisms.However, the original Transformer remained fundamentally a model designed for sequence data. This limitation naturally sparked an important line of thinking within the research c...
Incremental Structure from Motion: Core Implementation Guide with Ceres Optimization
Introduction: Building on Simulation FoundationsIn the previous installment of this series, we completed a crucial preparatory step—constructing a "virtual world" that adheres to strict physical optics laws and photogrammetry specifications. By simulating a UAV aerial survey workflow, we generated a synthetic dataset containing real terrain (DEM), camera poses, and feature point observations (tracks). This dataset grants us a "god's eye view": before the algorithm even runs, we already know the ground truth of the reconstruction.With reliabl...
Codex vs OpenClaw: Understanding the Fundamental Differences in AI Agent Architecture
Introduction: Navigating the AI Agent Terminology LandscapeIn recent months, practitioners across the industry have increasingly encountered terms like Codex, OpenClaw, MCP, A2A, Skill, and Harness. These concepts frequently appear together, and all relate to AI Agents, leading many to initially think: "They seem similar, but I can't quite pinpoint the differences."This article aims to resolve that confusion by explaining these easily mixed-up concepts within a unified framework, using accessible language that demystifies the abstract termin...
Mastering CLAUDE.md: The Complete Guide to Project Memory and Team Collaboration
The Problem: Why Claude Code Keeps Forgetting Your RulesImagine this scenario: It's Monday morning, and you fire up Claude Code to finish a feature you started last week. You type in your request: "Add a batch query method to OrderService that returns order details based on a list of order IDs."Claude quickly generates code. But as you scan it, your frown deepens—it's using a basic ArrayList for-loop with individual SQL queries, one at a time. Yet your project already has a well-encapsulated batchQuery() method ready to use.You correct it: "...