Articles
Field notes.
Answers to what people running Claude Code, Codex and other agents ask. Each one is filed under the words they search for.
-
claude code notification when waiting for input
Claude Code notifications: know when it’s waiting for input, even on your phone
A permission prompt sat for 20 minutes while you worked in another tab. Three fixes, from a one-line hook to your iPhone’s Lock Screen, and what each one misses.
Read -
claude code vs codex
Claude Code vs Codex on one codebase: 9,392 commits in 17 days
Not a benchmark. 490 sessions of Claude Code and Codex built one Mac and iPhone app on one branch, under one rule file. What the git history says about each.
Read -
tool to manage multiple claude code sessions
How to manage multiple Claude Code sessions without losing track
Past four or five sessions, tabs and tmux stop telling you who is working, who is stuck and who finished an hour ago. Four habits that hold, and one list that does the watching.
Read -
use claude code and codex together
How to use Claude Code and Codex together: one writes, the other reviews
One rules file both agents read, one worktree per agent that edits, and Codex reviewing Claude Code’s diff before you merge, or the other way round. Every command run with Claude Code 2.1.330 and Codex CLI 0.171.0.
Read -
claude code close laptop
Keep Claude Code running when you close your laptop: three ways
Close the lid and the agent stops. Keep the Mac awake, rent a server and tmux in, or hand the job to a Cloud computer: what each costs, and what each still leaves to you.
Read -
dangerously skip permissions vs auto mode
Claude Code’s --dangerously-skip-permissions vs auto mode, and the settings in between
What each Claude Code permission mode lets through, what auto mode still blocks, and the rules, sandbox and overnight settings that let routine work run while the irreversible steps wait for you.
Read -
claude md best practices verify coding agents
Proof, not claims: six verification rules from 17,468 AI agent commits
Six rules for CLAUDE.md and AGENTS.md from 17,468 commits by coding agents in 30 days, each with the measured failure that forced it, ready to paste into your own.
Read -
best app to run claude code and codex
Every app that runs Claude Code and Codex in one place, compared (November 2026)
Amia against T3 Code, Orca, herdr, Happy, Paradev, Rex and Pushary, in one chart: which agents each runs, where they keep working when your laptop shuts, what your phone can do, and what you pay. Only Amia works over every app on your Mac and pays and signs in without seeing your card or password. Read from each product’s own site and docs on 21 November 2026.
Read -
t3 code alternative
Amia vs T3 Code, Orca, herdr and Happy: what the free agent apps leave you to run
Amia, T3 Code, Orca, herdr and Happy all put Claude Code and Codex side by side. Only Amia works over every app on your Mac, pays and signs in without seeing your card or password, and comes with a Cloud computer and a private network, so there is no server to keep. Read from each project’s own site, docs and README on 21 November 2026.
Read
Hand it off tonight.
Free with your own AI accounts · Plans from … a month