
Security News
Agent Baseline Publishes Security Reference Architecture for Software Factories
Agent Baseline’s new reference architecture shows how to secure software factories with four boundaries enforced outside the AI agent.
@maya4art/builder-codex
Advanced tools
Solo-survival + behavioral guardrails standards bundle for AI-assisted development. Installs AGENTS.md, BUILDER_CODEX.md, skills/, and supporting files into your project root.
Standards bundle สำหรับ solo builder ที่ใช้ AI 100% ในการเขียนโค้ด v3.1.0 · 2026-06-22 · MIT-style license
| สถานการณ์ | เปิดอ่าน |
|---|---|
| 🆕 เริ่ม project ใหม่ | AGENTS.md (copy ไป root ของ project) |
| 🤖 ตั้ง AI agent ในเครื่องใหม่ | AGENTS.md (auto-loaded) + linked BUILDER_CODEX.md |
| 🐛 เจอ bug ที่ไม่รู้แก้ยังไง | BUILDER_CODEX.md § Pattern Library + § Debugging Toolkit |
| 📋 ก่อน deploy | BUILDER_CODEX.md § Pre-Deployment Checklist |
| 🔍 AI generate code แล้วไม่แน่ใจ | skills/verifying-ai-output/SKILL.md |
| 💾 ตั้งระบบ backup | skills/backing-up-solo-projects/SKILL.md |
| 📱 ก่อน submit Play/App Store | skills/preparing-mobile-store-submission/SKILL.md |
| 🇹🇭 ก่อน collect user data ใน Thai | skills/complying-with-thai-pdpa/SKILL.md |
| ✍️ เริ่ม feature ใหญ่ | skills/writing-feature-specs/SKILL.md |
| 🆕 AI ทำผิดอะไร — log อะไร | MEMORY.md (AI เขียนเอง) |
| 🆕 ตอนนี้ project อยู่ตรงไหน | spec.md § Current State |
| 🆕 R0/R1/R2 จัดยังไง | BUILDER_CODEX.md § P-31 |
your-project/
├── AGENTS.md ← Tier 1: <150 lines, agent auto-loads
├── @MEMORY.md ← 🆕 v3.1: AI's failure log (auto-imported)
├── @CAPTURE_LOG.md ← Human's strategic log (auto-imported)
├── spec.md ← 🆕 v3.1: Project save point
├── BUILDER_CODEX.md ← Tier 3: master reference (~700 lines)
├── DR_PLAN.md ← Disaster recovery (7 scenarios)
├── skills/ ← Tier 2: on-demand load
│ ├── verifying-ai-output/SKILL.md
│ ├── backing-up-solo-projects/SKILL.md
│ ├── preparing-mobile-store-submission/SKILL.md
│ ├── complying-with-thai-pdpa/SKILL.md
│ └── writing-feature-specs/SKILL.md
└── features/ ← Per-feature artifacts (P-29)
└── XYZ-feature-name/
├── SPEC.md ← ≠ project-level spec.md
├── PLAN.md
├── TASKS.md
└── REVIEW.md
Adopted from somnus0x/agt-skill-pack/claude-md-setup:
# 1. Copy bundle to project root
cd ~/projects/your-project
cp -r ~/builder-codex/AGENTS.md .
cp -r ~/builder-codex/BUILDER_CODEX.md .
cp -r ~/builder-codex/MEMORY.md .
cp -r ~/builder-codex/spec.md .
cp -r ~/builder-codex/CAPTURE_LOG.md .
cp -r ~/builder-codex/DR_PLAN.md .
cp -r ~/builder-codex/skills .
# 2. Customize 5 files (search for [Project Name] and replace)
# - AGENTS.md (stack + commands)
# - MEMORY.md (project name)
# - spec.md (project name + goal)
# - CAPTURE_LOG.md (project name)
# - DR_PLAN.md (project name + inventory)
# 3. Commit
git add .
git commit -m "chore: install Builder Codex v3.1.0"
# 4. AI agent (Claude Code, Cursor, Codex CLI, Windsurf, etc.) จะอ่าน AGENTS.md
# + imported MEMORY.md + CAPTURE_LOG.md อัตโนมัติทุก session
Claude Code, Cursor, Codex CLI, GitHub Copilot, Windsurf, Amp, Devin, Aider, Zed, Jules, VS Code, JetBrains Junie, Gemini CLI (ผ่าน symlink GEMINI.md → AGENTS.md)
@import syntax in AGENTS.mdเอกสารนี้เปิด — fork, modify, redistribute ได้
กฎเดียว: อย่าให้ AI generate "best practices" ทั่วไปมาใส่ — ทุก pattern ต้องมาจากบทเรียนจริง (ดู §วิธีใช้ ใน BUILDER_CODEX)
Synthesized from:
somnus0x/agt-skill-pack/claude-md-setup (behavioral guardrails layer)Original author: Tony (solo builder, Thailand) License: MIT-style — use freely, attribution appreciated
🛠️ Stay safe, ship reliable, let nothing fail silently. Compound, don't repeat.
FAQs
Solo-survival + behavioral guardrails standards bundle for AI-assisted development. Installs AGENTS.md, BUILDER_CODEX.md, skills/, and supporting files into your project root.
We found that @maya4art/builder-codex demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Security News
Agent Baseline’s new reference architecture shows how to secure software factories with four boundaries enforced outside the AI agent.

Security News
arXiv now limits authors to two submissions a month as AI slop overwhelms moderators, delays good papers, and sparks debate over applying the limit to everyone.

Research
/Security News
A new GhostAction wave hits hundreds of GitHub repos, expanding CI/CD secret theft to cloud and AI credentials in source code and git history.