#Developer Experience
3 posts tagged with "Developer Experience"
Building AI Agents That Refuse to Guess
An agent that can take action is not automatically safe to trust. These are the patterns I use to make coding agents inspect first, ask for intent, gate side effects, and stop when they lack confirmation.
Best Practices for Writing Agent Skills
I've been writing agent skills for Claude Code, these are the patterns that held up: thin SKILL.md files, deterministic scripts for the boring parts, progressive disclosure, and being disciplined about how many skills you leave enabled.
How I Actually Work With AI (OpenCode, Antigravity, Claude, and Codex)
A research-backed breakdown of my real AI workflow across OpenCode, Antigravity, Claude, and Codex, including models, skills, MCPs, and how I orchestrate them in practice.