Author
I gave one nasty order-dependent flaky test to Claude Code, Cursor, Codex CLI, Cline, and Gemini CLI. Here is which AI coding agent actually debugs in 2026, and when to reach for each.
I ran one real refactor through Claude Code, Cursor, Aider, and Codex CLI, ranked by safety net and review surface, not raw model smarts. A 2026 field log.
I ran the same webhook-handler feature through Gemini CLI and Claude Code on one repo. Gemini CLI wins for large-repo exploration and giant context; Claude Code wins for consistent multi-file changes. A 2026 field log with my routing rule, corrected after Google withdrew Gemini CLI's free individual tier.
OpenCode is open source and model-agnostic; Claude Code is managed and tuned around one vendor. I ran the same invite feature through both on the same repo in 2026. Here is which I kept, and the routing rule I use now.
A three-month field log on the best MCP servers in 2026: the five that earned a permanent slot, the ones I uninstalled, and the context and trust costs no listicle mentions.
My CI step running claude -p hung for the full timeout, waiting on a permission prompt no human would answer. The three flags that fix headless Claude Code, the jobs I kept, and the two I ripped back out.
A field log of Claude Code's built-in /sandbox: what its OS-level cage actually locks down, the setup that bit me, what auto-allow still stops for, the tools that broke, and the one thing it does not protect.
A first-person weekend log of cloning and running ai-app-builder-open, an open source AI app builder, on localhost. What worked, the one API key you need, and the honest catch versus local-first Dyad.
Two weeks of pushing real work onto Claude Code subagents. The two I kept, the frontmatter line that actually decides delegation, and the gotcha that cost me an afternoon.
I ran spec-driven development for a month on a live B2B product. An honest 2026 field log: where the spec earned its overhead, where it was pure ceremony, and the six-heading spec that survived.
Claude Code can now drive a real browser: read the console, click through your app, fill forms, pull data. Two weeks of using it on live work, what it replaced, where it still hands the keyboard back, and the context-cost footgun nobody warns you about.
The standalone /output-style command is gone as of v2.1.91, so most guides are stale. Here is the current way to switch styles, which of the four built-ins earn their place, and the keep-coding-instructions flag that quietly breaks custom styles.
