https://cubxxw.com/tags/harness-engineering/ 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 Sat, 03 Oct 2026 00:15:00 +0800 https://cubxxw.com/ai-agent/posts/personal-agent-harness-openclaw/ Sat, 03 Oct 2026 00:15:00 +0800 Sat, 03 Oct 2026 00:15:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/personal-agent-harness-openclaw/ Using OpenClaw's October 2026 source snapshot, this article studies how agent harnesses make reliable multi-day Personal Agent work and Computer Use possible. AI Agent Harness Engineering Automation Security Development Open Source Concept illustration: a Personal Agent's bounded run, persistent task, execution environment and result receipt; not an accurate OpenClaw architecture diagram https://cubxxw.com/ai-agent/posts/agent-system-design-openclaw/ Fri, 07 Aug 2026 17:40:00 +0800 Fri, 07 Aug 2026 17:40:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/agent-system-design-openclaw/ Trace OpenClaw v2026.7.1-2 through admission, bindings, agents and session keys to see how identity dimensions shape continuity, isolation and permissions. AI Agent Harness Engineering Security Monitoring Automation Development OpenClaw routing from internet channels through the resident Gateway, deterministic bindings and session keys into per-agent enclaves, device nodes and external side effects https://cubxxw.com/ai-agent/posts/agent-system-design-n8n/ Fri, 07 Aug 2026 17:25:00 +0800 Fri, 07 Aug 2026 17:25:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/agent-system-design-n8n/ Trace n8n 2.33.6 across Redis queues, workers and Postgres to see why scaling workflow throughput does not guarantee exactly-once effects in external systems. AI Agent Automation Harness Engineering Monitoring Security Development The n8n deterministic workflow spine, the constrained Agent spur, human and idempotency gates, external receipt reconciliation, and the Redis worker Postgres queue-mode yard https://cubxxw.com/ai-agent/posts/agent-system-design-pi/ Fri, 07 Aug 2026 16:40:00 +0800 Fri, 07 Aug 2026 16:40:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/ai-agent/posts/agent-system-design-pi/ Trace Pi v0.84.1's agent loop, default tools and JSONL sessions to see what a minimal kernel keeps and who takes over security, recovery and workflow decisions. AI Agent Harness Engineering Open Source Development Security Context Engineering The Pi minimal agent kernel, the detachable extension rail, provider and host OS trust boundaries, and the JSONL session tree below 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/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-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/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/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/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/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/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/growth/posts/super-individual-ai-product-and-solo-builder-stack/ Wed, 24 Jun 2026 14:55:00 +0800 Wed, 24 Jun 2026 14:55:00 +0800 熊鑫伟 (Xinwei Xiong) https://cubxxw.com/growth/posts/super-individual-ai-product-and-solo-builder-stack/ As the model layer commoditizes and 40% of enterprise AI projects are headed for cancellation, the super-individual finally has the full primitive set to cut into a $4.6T services market. This essay assembles the most current 2025-2026 methodologies into one executable map: the core thesis, six product directions, five ops stacks, the Soul Core + Harness Engineering foundations, and a 12-month roadmap. Super Individual Solo Builder AI Agent Harness Engineering MCP Product Strategy Personal Growth The Super-Individual Stack: AI-Native Product Directions and Solo Builder Ops in 2026 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