Giving AI Coding CLIs Full Permission to Run Autonomously - Claude Code, Codex, and Gemini CLI Compared

To let AI coding CLIs run autonomously, you need to configure their permission settings. Claude Code has --dangerously-skip-permissions and auto mode. Codex CLI has --full-auto and --yolo. Gemini CLI has --approval-mode yolo. This post compares the three tools' permission models and explains which mode to use in which situation.

Should Every Service Ship a Claude Code Plugin? — Rethinking Distribution After codex-plugin-cc

OpenAI's official codex-plugin-cc is genuinely good: I never had to leave Claude Code to drive Codex reviews and investigations. Plugins are working as a distribution channel that delivers a service inside the surface developers already use. Starting from codex-plugin-cc's design, this post discusses which services should ship plugins from day one, and the tradeoffs to weigh before committing.

Comparing Multi-Platform Strategies: Claude Code, Codex, and Gemini

The three major AI coding tool providers, OpenAI Codex, Claude Code, and Google Gemini, have each solved the multi-platform problem differently. Codex uses a bidirectional JSON-RPC protocol (App Server), Claude Code exposes its harness as a library (Agent SDK), and Google fields three separate products along an autonomy spectrum. This post compares their approaches and extracts patterns for designing your own multi-platform product.

Build a Documentation Site on Private GitHub Pages from Day 1

Documentation tends to get deprioritized, but setting up a documentation site on Day 1 makes sharing setup guides, architecture decisions, and operational procedures dramatically easier. With Private GitHub Pages, only team members with repository read access can view the site. This post introduces VitePress and Starlight as two fast-building options for internal documentation.

Why I Quit Pomodoro in the AI Era - Let AI Run on Autopilot While Humans Rest

25 minutes on, 5 minutes off. I quit Pomodoro after years of faithful practice. When AI agents can work autonomously for hours, a fixed 25-minute timer makes no sense. My new rhythm: let AI work, and let the human rest.