https://cubxxw.com/ai-agent/ Xinwei Xiong's open archive of AI workflows, real projects, engineering methods, systems thinking, personal growth, philosophy, and life. https://cubxxw.com/assets/og-image.png https://cubxxw.com/ Hugo -- gohugo.io en-us Wed, 05 Aug 2026 00:00:00 +0800 https://cubxxw.com/ai-agent/posts/claude-tag-organizational-agent-runtime/ Wed, 05 Aug 2026 00:00:00 +0800 Wed, 05 Aug 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/claude-tag-organizational-agent-runtime/ Claude Tag turns Slack into a shared agent runtime with identity, memory, and tools. This deep dive covers its architecture, risks, Chinese peers, and future. AI Agent Automation Context Engineering Product Strategy Security Productivity LLM A shared organizational agent dispatch console connected to permission vaults, memory archives, and isolated work cells https://cubxxw.com/ai-agent/posts/common-sense-ai-industry-reconstruction/ Sat, 01 Aug 2026 23:00:47 +0800 Sat, 01 Aug 2026 23:00:47 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/common-sense-ai-industry-reconstruction/ A framework for testing AI industry ideas: start with durable demand, map constraints and scarcity, redesign the workflow, and validate it with real failures. AI Agent Product Strategy Harness Engineering Automation Career A fixed demand axis beside industry workflow cards being reconnected https://cubxxw.com/ai-agent/posts/2026-07-31-forgetting-is-an-ai-system-capability/ Fri, 31 Jul 2026 01:10:00 +0800 Fri, 31 Jul 2026 01:10:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/2026-07-31-forgetting-is-an-ai-system-capability/ Cheap storage makes forgetting a core AI capability. This essay designs auditable decay, archiving, and revocation for long-running AI memory systems. Technology AI Agent Context Engineering Self-Discovery Philosophy Personal Growth Hierarchical memory systems heat active nodes, sink outdated information, and retain audit trails https://cubxxw.com/ai-agent/posts/open-design/ Wed, 22 Jul 2026 20:30:00 +0800 Fri, 31 Jul 2026 16:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/open-design/ A version-pinned guide to Open Design 0.16.1: its four-plane architecture, local and cloud paths, BYOK privacy boundaries, SSRF controls, and practical fit. Development AI Agent LLM Project Learning MCP Open Source Open Design as a four-plane design harness for coding agents https://cubxxw.com/ai-agent/posts/claude-code-boris-121-tips-playbook/ Mon, 20 Jul 2026 23:30:00 +0800 Mon, 20 Jul 2026 23:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/claude-code-boris-121-tips-playbook/ A practical Claude Code guide to verification, permissions, worktrees, subagents, workflows, goals, loops, and routines, with current limits and safe defaults. Development AI Agent Automation Harness Engineering Productivity Testing Claude Code playbook for verification, permissions, worktrees, loops, and parallel agents https://cubxxw.com/ai-agent/posts/prompt-loop-engineering-practice/ Mon, 20 Jul 2026 21:00:00 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/prompt-loop-engineering-practice/ Loop engineering turns repeated prompting into a verifiable system. A field guide to state, layered checks, permissions, and clear stopping rules for solo work. Development AI Agent LLM Automation Harness Engineering Testing Solo Builder A quiet control desk representing verifiable loop engineering for solo builders https://cubxxw.com/ai-agent/posts/solo-creator-video-pipeline/ Sun, 19 Jul 2026 22:00:00 +0800 Fri, 31 Jul 2026 20:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/solo-creator-video-pipeline/ A practical AI video editing pipeline for solo creators: structure shoots, automate rough cuts, keep judgment human, and publish safely across platforms. Development AI Automation Super Individual Content Strategy Solo Builder Harness Engineering A solo creator's AI video editing pipeline from camera to publication https://cubxxw.com/ai-agent/posts/super-individual-stack-reputation/ Sun, 19 Jul 2026 11:00:00 +0800 Sun, 19 Jul 2026 11:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-stack-reputation/ When the cost of building approaches zero, having built something no longer proves anything, so people go look at who built it. This essay covers the first layer of the gear stack — reputation: why trust is the one thing AI cannot flatten, the three carriers it lives on, why it will never appear on any tool list, and its three most uncomfortable truths. It closes by answering the question the whole series has been building toward: how to wire production, judgment, distribution, and reputation into one loop that feeds itself, instead of four unrelated chores. AI Super Individual Open Source Solo Builder Personal Growth Product Strategy When Anyone Can Build the Thing, "He Built It" Becomes the Signal https://cubxxw.com/ai-agent/posts/super-individual-stack-distribution/ Sun, 19 Jul 2026 10:30:00 +0800 Sat, 01 Aug 2026 03:35:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-stack-distribution/ A distribution strategy for solo builders: turn product work into durable content, choose channels deliberately, and measure the audience you actually own. Development AI Super Individual Content Strategy GEO Solo Builder Product Strategy Distribution for solo builders: a durable audience outlasting copied products https://cubxxw.com/ai-agent/posts/super-individual-stack-judgment/ Sun, 19 Jul 2026 10:00:00 +0800 Sun, 19 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-stack-judgment/ When execution cost approaches zero, the price of building the wrong thing doesn't shrink — it becomes payable in full. AI generates requirements at near-zero cost, so your queue fills with tasks that look reasonable and deliver little, and agents will faithfully complete every one. This essay takes apart the three jobs of the judgment layer: how to guard the requirement gate, how to write acceptance criteria, and how to externalize judgment into rules files the agent reads on every run — the only part of this layer that compounds. It's also honest about the half of judgment that rules can't hold, and why hit rate is the only real test of judgment. AI Agent Super Individual Context Engineering Product Strategy Solo Builder AI Writes Requirements for Free, and That's the Most Dangerous Part https://cubxxw.com/ai-agent/posts/super-individual-stack-production/ Sun, 19 Jul 2026 09:30:00 +0800 Sun, 19 Jul 2026 09:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-stack-production/ Part two of "The Super Individual's Gear Stack," on the bottom layer — production. Every advance in this layer is a public good, handed to everyone at the same moment, so once it's good enough you should stop investing. What actually blocks you isn't execution bandwidth but review bandwidth, which cannot be scaled the same way. The only lever that works is raising first-pass correctness — front-loading rules, writing acceptance criteria into tickets, and letting agents run tests to green before opening a PR. The back half lays out empirical data on AI code quality and security, which happens to explain exactly why testing and isolation are the real gear of this layer. AI Agent Super Individual Harness Engineering Testing Security I Ran Ten Agents Overnight, Woke Up to Ten PRs, and Then I Got Stuck https://cubxxw.com/ai-agent/posts/super-individual-stack-four-layers/ Sun, 19 Jul 2026 09:00:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-stack-four-layers/ A four-layer AI stack for super individuals: separate production tools from judgment, distribution, and reputation assets that compound over time in practice. Development AI Agent Super Individual Solo Builder Product Strategy Productivity Four-layer stack rising from AI production tools through judgment and distribution to reputation https://cubxxw.com/ai-agent/posts/devbox-doctor-design/ Sat, 18 Jul 2026 13:30:00 +0800 Sat, 18 Jul 2026 13:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/devbox-doctor-design/ A design for auditing Mac developer tools with read-only evidence, bounded AI judgments, privacy safeguards, graceful fallbacks, and false-positive tests. Development AI LLM Agent Context Engineering Automation DevOps Designing devbox-doctor, a safer Mac toolchain audit https://cubxxw.com/ai-agent/posts/designing-valuable-agent-skills/ Sat, 18 Jul 2026 00:30:00 +0800 Sat, 18 Jul 2026 00:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/designing-valuable-agent-skills/ Agent Skill and SKILL.md design through a storage-cleanup case study: architecture, permission boundaries, prompt contracts, real-machine tests, and lessons. Development AI LLM Agent Context Engineering Automation Security Agent Skill design shown as code, model judgment, permission gates, and human confirmation https://cubxxw.com/ai-agent/posts/info-to-creation-arsenal/ Fri, 17 Jul 2026 21:30:00 +0800 Fri, 17 Jul 2026 21:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-arsenal/ Build an evidence-driven AI knowledge base with executable workflows, method cards, voice rules, review gates, and feedback loops that improve content work. Development AI LLM Agent Context Engineering Personal Growth Automation A private AI knowledge arsenal connecting evidence cards, executable workflows, publishing platforms, and feedback loops https://cubxxw.com/ai-agent/posts/ai-native-blog-rebuild/ Fri, 17 Jul 2026 17:30:00 +0800 Fri, 17 Jul 2026 17:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-native-blog-rebuild/ I rebuilt a 120-post Hugo blog with AI agents, human review, MCP, GitHub Actions, a cover pipeline, and a measured GEO experiment. Here is what held up. Technology AI Agent MCP Automation Harness Engineering GEO Blog How AI agents rebuilt a 120-post Hugo blog with a human editor https://cubxxw.com/ai-agent/posts/ai-agent-red-ocean-blue-ocean-2026/ Wed, 15 Jul 2026 18:00:00 +0800 Wed, 15 Jul 2026 18:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-agent-red-ocean-blue-ocean-2026/ A practical map of defensible AI agent businesses: vertical workflows, accountable delivery, agent infrastructure, and moats that survive model progress. Development AI LLM Agent Product Strategy Solo Builder Super Individual A narrow blue passage beyond a crowded red ocean, representing defensible AI agent businesses https://cubxxw.com/ai-agent/posts/trusting-unattended-ai-agent/ Wed, 15 Jul 2026 16:00:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/trusting-unattended-ai-agent/ A practical trust stack for unattended AI agents: hard tool controls, regression evals, checkpoints, rollback, and focused human review for risky actions. Development AI LLM Agent Harness Engineering Monitoring Security Testing A quiet control room supervising an unattended AI agent workflow https://cubxxw.com/ai-agent/posts/super-individual-intelligence-system/ Wed, 15 Jul 2026 14:30:00 +0800 Fri, 31 Jul 2026 23:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/super-individual-intelligence-system/ Build a personal intelligence system that turns monitored sources into decisions, with honest boundaries for RSS, RSSHub, agents, MCP, safety, and evaluation. Development AI Agent MCP Automation Super Individual Context Engineering Monitoring A personal intelligence pipeline connecting monitored sources to evidence, judgment, and action https://cubxxw.com/ai-agent/posts/open-model-cost-collapse-agent-fleet/ Wed, 15 Jul 2026 14:00:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/open-model-cost-collapse-agent-fleet/ A dated, reproducible agent-fleet cost model that tests low-cost APIs, weighs open-weight options, and routes work by measured success and operational risk. Development AI LLM Agent Open Source Super Individual Solo Builder A measured cost test for routing work across an agent fleet https://cubxxw.com/ai-agent/posts/proactive-agent-it-prompts-you/ Wed, 15 Jul 2026 12:00:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/proactive-agent-it-prompts-you/ A practical framework for proactive AI agents: memory, runtimes, triggers, interruption budgets, human approval, and the discipline to stay quiet by default. Development AI LLM Agent Harness Engineering Product Strategy Context Engineering When the AI Agent Starts Prompting You, What Has Actually Changed https://cubxxw.com/ai-agent/posts/ai-auto-news-pipeline-limits/ Wed, 15 Jul 2026 10:00:00 +0800 Wed, 15 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-auto-news-pipeline-limits/ AI news pipelines can collect, translate, summarize, and deduplicate at scale. This field-tested guide shows where automation ends and human judgment begins. Development AI LLM Agent Automation Super Individual Monitoring An AI news pipeline reaching the boundary between automation and human judgment https://cubxxw.com/ai-agent/posts/from-chatbot-to-agent-to-skill/ Tue, 14 Jul 2026 10:00:00 +0800 Tue, 14 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/from-chatbot-to-agent-to-skill/ A practical Chatbot-to-Agent-to-Skill framework for turning domain judgment into reusable workflows, with Amazon Ads boundaries, metrics, and human approval. Development AI LLM Agent Context Engineering Product Strategy Automation Chatbot to Agent to Skill, a three-stage framework for reusable AI workflows https://cubxxw.com/ai-agent/posts/give-ai-tasks-not-directions/ Sat, 11 Jul 2026 16:30:00 +0800 Sat, 11 Jul 2026 16:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/give-ai-tasks-not-directions/ Explore direction with AI, then define the finish line yourself. A six-part task card turns open-ended conversation into work you can test, review, and trust. Development AI LLM Agent Context Engineering Productivity Learning A six-part AI task card connecting an open direction to a verifiable result https://cubxxw.com/ai-agent/posts/engineering-discipline-ai-workflow/ Sat, 11 Jul 2026 15:30:00 +0800 Sat, 11 Jul 2026 15:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/engineering-discipline-ai-workflow/ A practical framework for reliable AI workflows: separate claims from logs, preserve evidence, run evals, set failure thresholds, and escalate risk to humans. Development AI LLM Agent Harness Engineering Testing DevOps Five quality gates surrounding an AI workflow, from evidence to human review https://cubxxw.com/ai-agent/posts/ai-second-brain-build/ Sat, 11 Jul 2026 15:00:00 +0800 Sat, 11 Jul 2026 15:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-second-brain-build/ Build a safe AI second brain with Obsidian and Claude: structure your vault, isolate sensitive notes, preview changes, back up data, test, and roll back. Development AI LLM Agent Automation Productivity A safe three-layer AI second brain built with Obsidian, Claude, and a capture inbox https://cubxxw.com/ai-agent/posts/info-to-creation-creation/ Sat, 11 Jul 2026 14:40:00 +0800 Sat, 11 Jul 2026 14:40:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-creation/ An AI content creation workflow for turning knowledge cards into audience-ready work, with evidence gates, human review, feedback, and measurable outcomes. Development Product Strategy Personal Growth AI LLM Self-Discovery Super Individual Automation Knowledge cards moving through an AI-assisted creation pipeline toward an audience, then returning as a measured feedback loop https://cubxxw.com/ai-agent/posts/info-to-creation-knowledge/ Sat, 11 Jul 2026 14:30:00 +0800 Sat, 11 Jul 2026 14:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-knowledge/ An AI knowledge base workflow that turns notes into verified, reusable capability through evidence gates, retrieval, review, and reversible retirement. Development AI LLM Agent Context Engineering RAG Automation Draft records passing an evidence gate into retrievable knowledge cards, then returning through a review and retirement loop https://cubxxw.com/ai-agent/posts/info-to-creation-record/ Sat, 11 Jul 2026 14:20:00 +0800 Sat, 11 Jul 2026 14:20:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-record/ An AI note-taking workflow for turning text, voice, screenshots, code, and decisions into traceable records that can be reviewed and promoted to knowledge. Development Personal Growth Self-Discovery AI LLM Automation Product Strategy Text, voice, screenshots, code changes, and decisions moving through capture, clarification, and review toward a verified knowledge card https://cubxxw.com/ai-agent/posts/info-to-creation-information/ Sat, 11 Jul 2026 14:10:00 +0800 Sat, 11 Jul 2026 14:10:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-information/ An AI information filtering workflow for capturing useful sources without building a noise archive, with privacy gates, verification, and human review. Development AI LLM Personal Growth Self-Discovery Product Strategy Automation Sources passing through relevance, privacy, and verification gates before AI-assisted processing and entry into the records layer https://cubxxw.com/ai-agent/posts/info-to-creation-the-framework/ Sat, 11 Jul 2026 14:00:00 +0800 Sat, 11 Jul 2026 14:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/info-to-creation-the-framework/ A practical AI knowledge workflow for moving information through traceable records and verified knowledge into audience-ready creation with human review. Development AI LLM Personal Growth Context Engineering Product Strategy Self-Discovery Automation Information crossing intake gates into records, verified knowledge, audience-facing creation, and a measured feedback loop https://cubxxw.com/ai-agent/posts/geo-measurement-and-tools/ Sat, 11 Jul 2026 12:00:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-measurement-and-tools/ Measure GEO without proxy myths: combine Google generative AI impressions, repeated citation audits, referral analytics, and conversions in one clear protocol. Development GEO SEO AI Search Content Strategy LLM GEO measurement framework connecting exposure, citations, referrals, and conversions https://cubxxw.com/ai-agent/posts/geo-blog-rebuild-case-study/ Sat, 11 Jul 2026 11:30:00 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-blog-rebuild-case-study/ A real-data GEO case study using Search Console and Lighthouse to separate noisy impressions from useful demand, protect a domain move, and plan a Hugo rebuild. Development GEO SEO AI Search Content Strategy Blog GEO blog rebuild case study cover showing a real-data dashboard and growth curve https://cubxxw.com/ai-agent/posts/geo-trust-and-endorsement/ Sat, 11 Jul 2026 11:00:00 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-trust-and-endorsement/ A practical 2026 guide to off-site GEO trust: author identity, third-party evidence, ethical community distribution, and repeatable citation measurement. Development GEO AI Search Content Strategy SEO AI LLM A verifiable network connecting an author, original work, independent evidence, and reader communities https://cubxxw.com/ai-agent/posts/geo-structured-content-tactics/ Sat, 11 Jul 2026 10:30:00 +0800 Fri, 31 Jul 2026 18:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-structured-content-tactics/ A practical guide to clearer technical articles: lead with useful answers, use honest evidence, add valid schema, build internal links, and test AI visibility. Development GEO Content Strategy SEO AI Search Hugo GEO structured tactics cover showing clear, reusable content blocks https://cubxxw.com/ai-agent/posts/geo-how-ai-retrieves-and-cites/ Sat, 11 Jul 2026 10:00:00 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-how-ai-retrieves-and-cites/ A practical 2026 guide to how Google, Perplexity, and ChatGPT retrieve and cite sources, separating official facts from inference and showing how to test GEO. Development GEO RAG AI Search LLM Content Strategy A testable model of AI search retrieval, source selection, answer use, and citations https://cubxxw.com/ai-agent/posts/geo-generative-engine-optimization-guide/ Fri, 10 Jul 2026 22:00:00 +0800 Fri, 31 Jul 2026 18:39:47 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/geo-generative-engine-optimization-guide/ An evidence-led guide to generative engine optimization in 2026: what Google, OpenAI, Perplexity, Pew, Bain, and the original GEO paper actually support. Development GEO SEO AI Search Content Strategy LLM A five-layer GEO workflow separating access, usefulness, evidence, extraction, and measurement https://cubxxw.com/ai-agent/posts/dissecting-open-lovable/ Mon, 29 Jun 2026 09:30:00 +0800 Mon, 29 Jun 2026 09:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/dissecting-open-lovable/ A commit-pinned audit of open-lovable’s text protocol, Agentic Search, Morph edit path, and E2B/Vercel sandbox boundaries for practical AI app builders. Development AI Agent LLM Harness Engineering Development Open Source open-lovable architecture from web scraping through code generation to dual-sandbox preview https://cubxxw.com/ai-agent/posts/relay-agent-architecture-design/ Wed, 24 Jun 2026 10:00:00 +0800 Fri, 31 Jul 2026 18:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/relay-agent-architecture-design/ A source-based audit of Relay’s coordinator, five domain agents, LangGraph HITL, browser delivery, durability gaps, and remaining 2026 migration debt. Development AI Agent LLM Python Security Development Editorial architecture diagram of Relay with a coordinator, five domain agents, safeguards, browser delivery, and audit boundaries https://cubxxw.com/ai-agent/posts/context-engineering-the-new-foundation/ Mon, 22 Jun 2026 03:30:00 +0800 Mon, 22 Jun 2026 03:30:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/context-engineering-the-new-foundation/ Learn how context engineering improves AI agents through retrieval, compaction, memory, and token budgeting—and why high-signal context beats a larger window. Development Context Engineering AI LLM Agent MCP A wide schematic of context engineering: the Write / Select / Compress / Isolate pillars feeding an AI, a laptop with notes, and a local-first world line https://cubxxw.com/ai-agent/posts/agent-engineering-the-98-percent-harness/ Wed, 17 Jun 2026 09:30:00 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/agent-engineering-the-98-percent-harness/ Agent Engineering Harness explained through eight production pillars: orchestration, context, memory, tools, reliability, evaluation, cost, and governance. AI Agent LLM Context Engineering Architecture MCP A technical diagram with a tiny agent loop at the center, surrounded by concentric rings of the eight pillars: orchestration, context, memory, tools, reliability, evaluation, cost, governance https://cubxxw.com/ai-agent/posts/agent-identity-from-locke-to-openclaw/ Sun, 05 Apr 2026 20:00:00 +0800 Sun, 05 Apr 2026 20:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/agent-identity-from-locke-to-openclaw/ A practical framework for continuous AI agent identity, covering memory, permissions, provenance, multi-agent handoffs, and cross-session evaluation design. Development AI Agent LLM Context Engineering Harness Engineering Philosophy Files, memory, runtime boundaries, and evaluation for continuous agent identity https://cubxxw.com/projects/mem0/ Fri, 09 May 2025 21:33:46 +0800 Fri, 09 May 2025 21:33:46 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/mem0/ A practical 2026 guide to Mem0 OSS v3: ADD-only extraction, hybrid retrieval, entity linking, governance, evaluation, and when not to use it in production. Development AI Agent Open Source Project Learning RAG LLM A quiet archive of linked memory cards illustrating Mem0 OSS v3 retrieval https://cubxxw.com/projects/UFO/ Fri, 09 May 2025 21:30:15 +0800 Thu, 30 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/UFO/ A practical analysis of Microsoft UFO³, tracing its path from Windows UI automation to Desktop AgentOS and cross-device orchestration, with limits and guidance. Development AI Agent Automation Development Open Source Microsoft UFO desktop agent architecture https://cubxxw.com/ai-agent/posts/ai-recommend/ Wed, 23 Apr 2025 10:39:04 +0800 Wed, 23 Apr 2025 10:39:04 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-recommend/ A practical guide to LLM recommendation systems: semantic retrieval, ranking, RAG, exploration, evaluation, feedback loops, and production trade-offs. Technology AI LLM RAG Monitoring Project Learning Blank cards passing through a wooden ranking staircase toward a reader https://cubxxw.com/projects/notebooklm/ Mon, 21 Apr 2025 22:59:57 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/notebooklm/ A practical 2026 guide to Gemini Notebook: grounded research, citations, Studio outputs, plan limits, privacy, mobile tradeoffs, and reliable workflows. Development AI LLM RAG Productivity Learning Context Engineering A research notebook connecting source pages, citations, and synthesized outputs https://cubxxw.com/projects/tdd/ Mon, 21 Apr 2025 15:52:34 +0800 Fri, 31 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/tdd/ A 2026 guide to test-driven development for AI and LLM systems, covering deterministic tests, evals, calibrated graders, CI risk gates, and monitoring. Development AI LLM Testing Python Go Automation A red, green, and blue testing loop surrounding an AI system https://cubxxw.com/projects/markitdown/ Mon, 21 Apr 2025 15:41:21 +0800 Mon, 21 Apr 2025 15:41:21 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/markitdown/ Learn MarkItDown 0.1.6 for document-to-Markdown workflows, choose local, OCR, or Azure conversion, and test output quality before building a RAG pipeline. Development AI Open Source Python LLM RAG Project Learning Documents flowing through a measured conversion pipeline into structured Markdown https://cubxxw.com/projects/langgraph/ Sat, 19 Apr 2025 15:19:20 +0800 Sat, 19 Apr 2025 15:19:20 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/langgraph/ A practical 2026 guide to LangGraph architecture: choose the right agent layer, build checkpointed StateGraphs, and recover safely from production failure. Development AI LLM Python Open Source Project Learning A checkpointed LangGraph StateGraph with recovery paths https://cubxxw.com/projects/langchain/ Wed, 16 Apr 2025 17:36:45 +0800 Fri, 31 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/projects/langchain/ A production-minded guide to LangChain 1.x: when to call a model directly, use create_agent, or adopt LangGraph, with tested patterns for RAG and HITL. Development AI Open Source LangChain Agent RAG Python LLM A restrained editorial illustration of connected paths representing models, agents, and graph workflows https://cubxxw.com/ai-agent/posts/ai-gateway/ Wed, 16 Apr 2025 17:36:12 +0800 Wed, 16 Apr 2025 17:36:12 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/ai-gateway/ A practical comparison of LiteLLM, Kong, APISIX, Cloudflare, and Portkey across routing, cost, security, deployment boundaries, and production checks. Technology AI LLM Cloud Native Security Monitoring Open Source DevOps Several model pipelines converge at an AI Gateway where cost and governance are weighed https://cubxxw.com/ai-agent/posts/independent-developer/ Tue, 15 Apr 2025 20:53:12 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/independent-developer/ A practical 2026 roadmap for solo builders: validate demand, choose a minimal stack, ship a maintainable MVP, charge early, and preserve a clear exit path. Development Solo Builder Product Strategy Development Git Docker Testing Deployment A solo developer's path from problem validation to a maintainable MVP https://cubxxw.com/ai-agent/posts/gpt-researcher/ Mon, 14 Apr 2025 16:17:27 +0800 Fri, 31 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/gpt-researcher/ A source-audited engineering guide to GPT Researcher: current Python APIs, Docker and MCP deployment, cost controls, benchmark limits, and production risks. Development AI Project Learning Open Source Python Docker MCP RAG GPT Researcher pipeline from query planning and retrieval to a cited report https://cubxxw.com/ai-agent/posts/jina/ Sat, 12 Apr 2025 13:01:59 +0800 Fri, 31 Jul 2026 12:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/jina/ A production-focused guide to Jina after its Elastic acquisition, covering v4/v5 embeddings, reranking, Reader, DeepSearch, MCP, licensing, and deployment. Development AI AI Search RAG Open Source Project Learning Cloud Native A layered search system connecting multimodal data, embeddings, reranking, and grounded answers https://cubxxw.com/ai-agent/posts/harnessing-language-model-applications-with-langchain-a-developer-is-guide/ Wed, 22 May 2024 21:37:34 +0800 Fri, 31 Jul 2026 10:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/harnessing-language-model-applications-with-langchain-a-developer-is-guide/ A practical LangChain 1.x guide to models, agents, middleware, memory, RAG, SQL, Ollama, and LangSmith, with runnable Python and safe migration advice. Development AI LLM Agent Python RAG Development Automation LangChain 1.x architecture connecting models, agents, memory, retrieval, and observability https://cubxxw.com/ai-agent/posts/exploring-large-language-models-llms-pioneering-ai-understanding-generation-human-language/ Wed, 15 May 2024 20:12:29 +0800 Wed, 15 May 2024 20:12:29 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/exploring-large-language-models-llms-pioneering-ai-understanding-generation-human-language/ Understand how Transformers, scaling, and RAG work—and how retrieval, citations, evaluation, security, and production rollout make LLM systems reliable. Development AI LLM RAG Context Engineering Development Testing Security From Language Models to RAG: Capabilities, Limits, and Engineering https://cubxxw.com/ai-agent/posts/sora-ease-guide-mastering-sora-ai-for-developers/ Thu, 14 Mar 2024 08:44:13 +0800 Fri, 31 Jul 2026 16:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/sora-ease-guide-mastering-sora-ai-for-developers/ A 2024 SoraEase prompt archive, updated after Sora’s shutdown, with a verified timeline and practical migration lessons for developers leaving the Videos API. Development AI Development Open Source Automation Testing Content Strategy An abandoned film strip crossing a bridge toward a new modular video workflow https://cubxxw.com/ai-agent/posts/exploring-sora-technology-for-enthusiasts-and-developers/ Sat, 24 Feb 2024 13:30:15 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/exploring-sora-technology-for-enthusiasts-and-developers/ A dated, evidence-based review of Sora, from its 2024 research preview and Sora 2 to shutdown, safety limits, copyright risk, and API migration for developers. Development AI Development Product Strategy Security Content Strategy A timeline of Sora from research preview and product launch to shutdown https://cubxxw.com/ai-agent/posts/vector-database-learning/ Sat, 20 Jan 2024 12:57:15 +0800 Sat, 20 Jan 2024 12:57:15 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/vector-database-learning/ An introduction to vector databases for AI applications. Starts from the prerequisites, the basics of vectors, similarity measures such as cosine similarity, and database indexing, then explains how vector databases differ from traditional databases and where they fit in AI. Blog AI https://cubxxw.com/ai-agent/posts/emerging-challenges-and-trends-in-2024/ Sun, 14 Jan 2024 22:52:24 +0800 Thu, 30 Jul 2026 22:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/emerging-challenges-and-trends-in-2024/ A 2026 reassessment of 2024 AI field notes, covering model emergence, Prefix LM, LoRA, QLoRA, agents, and the engineering required for reliable RAG today. Development AI LLM RAG Agent Monitoring From 2024 AI field notes to a reliable production RAG system https://cubxxw.com/ai-agent/posts/use-auto-gpt/ Sat, 18 Mar 2023 16:28:30 +0800 Fri, 31 Jul 2026 00:00:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/use-auto-gpt/ Learn what changed between AutoGPT Classic and the 2026 AutoGPT Platform, why the old tutorial is unsafe, and how to migrate workflows with clear controls. Development AI Agent Automation LLM Python Docker Open Source A clockwork cart passing through controlled gates, representing the evolution from AutoGPT Classic to a governed agent platform