{"slug": "how-i-built-a-production-claude-code-setup-and-open-sourced-it", "title": "How I Built a Production Claude Code Setup (and Open-Sourced It)", "summary": "Developer Faiz Khairi open-sourced Claude Code Blueprint, a library of configuration files that improve the reliability and consistency of Anthropic's Claude Code AI coding assistant. The blueprint includes 12 agents, 17 skills, 12 hooks, and 6 rules designed to prevent common AI coding mistakes and add persistent memory between sessions. It is MIT-licensed and available on GitHub.", "body_md": "I have been using Claude Code seriously for months. And I kept hitting the same frustrations:\n\nSo I built a reference architecture that fixes these. Then I open-sourced it.\n\n[Claude Code Blueprint](https://github.com/faizkhairi/claude-code-blueprint) is a library of ready-to-copy files: CLAUDE.md, hooks, agents, skills, and rules. You copy what you need into your own project. Nothing more.\n\nIt is not a framework. It is not a wrapper. It is configuration files that change how Claude Code behaves.\n\n**12 agents * 17 skills * 12 hooks * 6 rules**\n\n60 seconds to start:\n\n`ash`\n\ncurl -o CLAUDE.md https://raw.githubusercontent.com/faizkhairi/claude-code-blueprint/main/CLAUDE.md\n\nThe CLAUDE.md ships with three behavioral rules that prevent the most common AI coding mistakes:\n\nBefore writing any fix, Claude must run four checks: git state, error source, existing suppression, minimum viable diagnosis. This stops the most common waste -- building an elaborate fix for a problem that was already solved in a previous commit.\n\nAny change touching more than one file goes through plan mode. No exceptions. Claude proposes, you approve, then it executes. This prevents the pattern where Claude charges into a 15-file refactor and produces something you did not ask for.\n\nAfter finishing any task, Claude runs a verification pass appropriate to the work type. API endpoint? It curls the live URL. Config change? It confirms the value was actually picked up. File edit? It re-reads the changed block. This alone catches a significant proportion of errors before you do.\n\nThis was the hardest part to get right. Claude Code has no native persistent memory between sessions. The blueprint solves this with a file-based memory system:\n\nEvery session starts with a load-session hook that reads these files. Every session ends with a save-session hook that writes them. The result: Claude picks up where it left off, even days later.\n\nOne of the things I learned the hard way: running everything on the expensive model wastes quota on mechanical work.\n\nThe blueprint ships with 12 agents, each pinned to the right model tier:\n\nThe main Claude Code session (Opus) orchestrates. Sonnet subagents do the bulk work. Your expensive quota goes to the decisions that need it.\n\nSkills are slash commands that invoke multi-step workflows. The blueprint ships 17:\n\n| Skill | What It Does |\n|---|---|\n| eview-full | Full PR review across security, logic, tests, style |\n| deploy-check | Pre-deploy checklist for your stack |\n| est-check | Run tests, check coverage, flag gaps |\n| sprint-plan | Turn a backlog into a sprint with estimates |\n| load-session | Restore context from memory at session start |\n| save-session | Write session state to memory at session end |\n| self-review | Review your own diff before pushing |\n\nHooks run outside Claude's context window. They cost zero tokens. They are the highest-leverage place to add guardrails.\n\nThe blueprint ships 12:\n\nAny developer using Claude Code who wants more reliable, consistent behavior. The blueprint is framework-agnostic -- it configures Claude's behavior, not your code.\n\nThere are four presets if you want a guided start:\n\n`ash`\n\n./setup.sh --preset=minimal # CLAUDE.md + 3 core hooks\n\n./setup.sh --preset=standard # + agents + skills\n\n./setup.sh --preset=full # everything\n\nThe README and setup docs are available in Japanese, Korean, and Chinese. The community contributed these.\n\nGitHub: [github.com/faizkhairi/claude-code-blueprint](https://github.com/faizkhairi/claude-code-blueprint)\n\nMIT licensed. Copy what you need. Leave the rest.\n\nIf it is useful, a star helps others find it.", "url": "https://wpnews.pro/news/how-i-built-a-production-claude-code-setup-and-open-sourced-it", "canonical_source": "https://dev.to/faizkhairi/how-i-built-a-production-claude-code-setup-and-open-sourced-it-1be4", "published_at": "2026-06-22 07:06:51+00:00", "updated_at": "2026-06-22 07:10:14.876171+00:00", "lang": "en", "topics": ["developer-tools", "large-language-models", "ai-tools", "ai-agents"], "entities": ["Faiz Khairi", "Claude Code", "Anthropic", "Claude Code Blueprint", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/how-i-built-a-production-claude-code-setup-and-open-sourced-it", "markdown": "https://wpnews.pro/news/how-i-built-a-production-claude-code-setup-and-open-sourced-it.md", "text": "https://wpnews.pro/news/how-i-built-a-production-claude-code-setup-and-open-sourced-it.txt", "jsonld": "https://wpnews.pro/news/how-i-built-a-production-claude-code-setup-and-open-sourced-it.jsonld"}}