How MCP Local File Access Works Now Since Roots Is Deprecated
How to give an MCP local file access under the stateless 2026-07-28 spec — the three sanctioned methods, the guardrails, and a token-efficient design.
Markdown vs HTML5 for AI: Write for Machines, Publish for Humans
I compared Markdown and HTML5 for AI workflows — token costs, processing accuracy, and human readability — plus which format your specs should use.
Migrate WordPress From Bitnami LAMP Stack Without Losing Your Settings
AWS deprecates Bitnami blueprints in November 2026. Here is how I moved six sites off Bitnami LAMP at full fidelity, and the seven things that broke on the way.
How to Easily Build a Custom GeneratePress 404 Page (No Premium Needed)
Build a custom GeneratePress 404 page with child theme code — a distinct title, search box, and recent posts — no Premium Elements or extra plugins needed.
7 Blogging Lessons I Learned From Building My Blog
Honest blogging lessons from 15+ years of building my blog — why you shouldn’t compete with the best, what it really costs, and why content alone won’t grow it.
Why WordPress Speed and SEO Are Inseparable
Core Web Vitals, crawl budget, bounce rates, and the rise of Answer Engine Optimization — here is why WordPress speed and SEO go hand in hand.
How to Optimize WordPress Category and Tag Pages for SEO
Turn WordPress category and tag pages into ranking landing pages. Learn descriptions, slug structure, noindex decisions, and real fixes from MetaBlogue.
AI Coding Ground Rules: What to Put in AGENTS.md First
AI coding ground rules tell your assistant what it can and can’t touch. Here are the rules I put in every AGENTS.md to stop scope creep and AI slop.
How to Set Up MCP Servers for Your AI Workflow
Learn how to set up MCP servers to connect your AI agent to web search, files, GitHub, and databases — plus a security checklist and a real content workflow.
Skills vs MCP: The Token Math That Changed My AI Setup
I compared token usage between converting documents to skills and running MCP servers. Here’s the math, and why the biggest savings sit outside the context window.
