Insights on AI, automation, and building the future of work

The stack that runs your work

The practice, rebuilt with AI
When AI can produce competent advice on almost anything, what still makes a human expert worth paying?

Four buckets, filed by your assistant

One is a method, one is the whole system

Cheaper than Claude Code, and it still forgets your project

The smart model writes. Your daily model keeps it.

6 free skills, blind-tested on Opus 4.8. Receipts included

Projects, Areas, Resources, Archives

The agent does the filing, not you

Map your setup, port it, then decide

Nine tools tested for executive-grade delegation

The practical June 2026 comparison for CLI, desktop, worktrees, hooks, MCP, and team rollout.

A practical June 2026 guide to Codex CLI for local repos, review, MCP, and scripted agent runs.

A June 2026 work-surface matrix for choosing the right AI agent across chat, code, desktop, IDE, cloud, and Google Antigravity.

What ContextBench, SWE-contextbench, AGENTS.md, and long-context studies actually test.

The useful ADHD AI assistant is the one that protects context and removes choices.

A 50-row feature matrix for Claude Code CLI, Desktop, Dispatch, hooks, MCP, and model support.

Seven team tools judged on AI-native recall

Seven assistants tested on real daily admin