Agent instructions, skills, SSH host aliases and expansion triggers are worth having identical on every machine one person owns, and belong in none of the shared configuration. They live in user/ now, with a manifest saying where each piece goes and a link-user stage that puts it there. That stage does nothing unless the machine said yes. Somebody who clones Panama to try the desktop keeps their own ~/.claude/CLAUDE.md exactly where it was; the question names the destinations and defaults to no. Anything displaced goes to config/old rather than being deleted. ~/.claude/CLAUDE.md and ~/.codex/AGENTS.md were byte-identical copies of one file, which is the drift this exists to prevent. Also adds the vitals toggles for the battery and Claude usage readouts, which had preferences and no way to reach them.
990 B
990 B
name, description, disable-model-invocation
| name | description | disable-model-invocation |
|---|---|---|
| wait-what | Stop and re-pitch the last message in plain English. | true |
Wait, What?
That last message did not land. Re-pitch it.
- Give enough context to make it land this time, rather than repeating the same words more slowly. Assume the gap was missing context, not missing intelligence.
- Write in Simplified Technical English: short sentences, one idea each, plain words, active voice, no jargon that has not been introduced.
- Use the project's ubiquitous language from
CONTEXT.md, followingCONTEXT-MAP.mdto the right one where the repo has several. Where a term is not in the glossary, define it in the sentence that first uses it. - Lead with the point. The re-pitch is not a longer version of the original: it is the same claim, arrived at from what the reader already knows.
Do not apologise, do not narrate the miscommunication, and do not simply expand the original. Say the thing again, better.