#contract drift
74 related articles
·3 min
Claude Code in Practice: Writing Maintainable AI Code with the Harness Methodology
Learn how to write controllable, maintainable AI code using the Harness methodology with Claude Code. Covers SDD, Agent orchestration, and enterprise-grade AI programming practices.
Read more →
·3 min
Why Claude Code Ditched RAG for Grep: The Engineering Logic Behind the Decision
Why did Claude Code abandon RAG for Grep? Breaking down the three root causes — undiagnosability, the multiplication effect, and index staleness — behind the shift to Agentic Search.
Read more →
·2 min
The Real Risk of AI Agent Deployment: Not Going Rogue, But Being Too Obedient
A real AI Agent customer service failure reveals: the biggest deployment risk isn't losing control — it's faithfully executing a poorly defined goal. A deep dive into Agent risks and boundary management.
Read more →
·6 min
Agent Studio: A Unified Platform for Authoring AI Roles and Skills
Agent Studio unifies AI Agent role definition (Subagents) and Skills on one platform, enabling coordinated orchestration through a shared MCP endpoint, progressive disclosure, and moderated community publishing.
Read more →
·7 min
Linked Context: How AI Agent Skills Can Point to the Entire Open Web
Deep dive into Skill Studio's "Linked Context" mechanism—skill files become real-time URL fetches instead of pre-loaded copies, extending AI Agent progressive disclosure to the entire open web.
Read more →
·3 min
Complete Guide to Harness Engineering: The Three-Layer Architecture for AI Engineering Development
A deep dive into Harness Engineering's core architecture covering the Information, Constraint, and Automation layers to systematically constrain and verify AI Agent output for reliable development.
Read more →
·2 min
Complete Guide to Claude Code Hooks: Deterministic Automation Configuration Explained
Deep dive into Claude Code Hooks: covering five event types, auto-formatting, dangerous operation blocking, and team collaboration best practices for deterministic AI coding.
Read more →
·4 min
From Vibe Coding to AI-Engineered Programming: A Practical Guide to Three Levels of Mastery
A deep dive into three levels of AI programming: Vibe Coding for rapid prototyping, Plan Mode for structured development, and AI-engineered programming for enterprise-grade projects with SDD and Claude Code SuperPower.
Read more →
·3 min
GLM-5.2 Deep Dive: Million-Token Context, MIT Open Source & Full-Stack Domestic Chip Training
Deep dive into Zhipu's GLM-5.2: truly usable 1M-token context, MIT open-source strategy, full-stack Huawei Ascend training, and how it compares to Claude Opus. Includes benchmarks, use cases & pricing.
Read more →
·4 min
Claude Code Codex Plugin Integration: A Practical Guide to Dual-AI Adversarial Review for Better Code Quality
Learn how to install and configure the Codex plugin in Claude Code, leveraging dual-AI adversarial review to uncover code vulnerabilities across seven attack surfaces.
Read more →
·4 min
Grill With Docs: Doubling AI Programming Collaboration Efficiency with a Unified Language
From Grill Me to Grill With Docs: applying DDD's Ubiquitous Language to solve terminology drift and context loss in AI programming for efficient cross-session collaboration.
Read more →
·2 min
Claude Code + Codex: The Most Reliable Division of Labor — A Three-Phase Approach of Planning, Execution, and Review
A detailed guide to the Claude Code and Codex collaborative workflow: Claude Code handles architecture planning and code review, while Codex handles execution. Boost dev efficiency through standardized handoffs.
Read more →
·3 min
Deep Dive into How Claude Code Works: Prompt Assembly, Permission Control, and the Skills Mechanism
Deep dive into Claude Code internals: stateless model principles, four-layer prompt assembly, Agentic Loop execution, permission control, and reusable Skills workflow templates for agentic engineering.
Read more →
Deep Dives·4 min
AI Hallucinations: Why Large Language Models Inevitably "Make Things Up" and How to Deal With It
Deep dive into AI hallucination's three root causes: training objective flaws, exposure bias, and probabilistic generation. Covers classification and practical mitigation strategies including RAG.
Read more →
Tutorials·3 min
Claude Code in Practice: A Real Development Workflow Completing a Full Feature in 18 Minutes
Real case study showing Claude Code's efficient dev workflow: four-part prompt template, three-round iterative fixes, code review methods, and decision frameworks to save 60% coding time.
Read more →
Deep Dives·3 min
How Multi-Agent Teams Solve AI Hallucination and Make AI Reliable
Deep analysis of how multi-agent architecture solves AI hallucination. From context rot to adversarial debate mechanisms, see how Anthropic, xAI, and Kimi reduce hallucination rates from 12% to 4.2%.
Read more →
Tutorials·3 min
7 Official Best Practices for Claude Code Recommended by Anthropic
Anthropic's official Claude Code best practices: 7 core tips covering self-review, plan mode, precise instructions, CLAUDE.md, reverse interviews, session cleanup, and multi-session parallelism.
Read more →
Product Reviews·2 min
ZenFlow Hands-On: Testing the Spec-Driven Fully Autonomous AI Software Engineer
Hands-on review of ZenFlow—the first spec-driven fully autonomous AI software engineer. Multi-agent parallel collaboration with built-in verification delivers end-to-end development from ideation to production.
Read more →
Tutorials·3 min
Harness Engineering: A Practical Guide to AI Development — From Prompts to Mastering AI Agents
A deep dive into Harness Engineering's three-layer architecture: Information, Constraint, and Automation layers, covering Agent failure modes, OpenAI and Anthropic best practices, and AI tool selection strategies for controlled AI development.
Read more →
Tutorials·3 min
Enterprise RAG Implementation: Architecture Principles and Production-Grade Optimization Guide
Complete guide to enterprise RAG architecture covering data indexing, vectorization, and retrieval optimization. Practical insights on chunking strategies, hybrid retrieval, and hallucination control for production-grade LLM applications.
Read more →