A context window isn’t memory. I use a three-layer Markdown setup, write-time checks, and regular archival so agents can pick up where yesterday left off.
read full log arrow_right_alt
Reusable Agent skills for spec, plan, build, test, review, and ship—bringing AI coding closer to real team workflows.
read full log arrow_right_alt
AI agents are shifting from one-shot prompts to feedback systems—verification, retry, state, and stop conditions form a reliable loop.
read full log arrow_right_alt
browser-use/browser-harness connects agents to real Chrome via CDP—for browsing, clicks, screenshots, forms, and reusable domain skills.
read full log arrow_right_alt