Claude Skills by w-is-h

What the brain holds and how it is kept — the agent's persistent memory of what is true. Load before restructuring or distilling it, or when unsure whether something belongs in the brain, a project's memory, or the notes.

The spec of memory — what belongs in a page and what stays out, how pages are keyed and nested, how a page gets created, links, the ceiling and what a refused write means. Load for any memory work beyond a one-line edit — a new page, restructuring, a cap refusal.

What the daily note is and how to write one. Load at the moment of writing or rewriting a note — the end of every session — or when the notes have drifted into a changelog or error ledger.

The end-of-session protocol — what gets written when the user signs off (\"bye\", \"good night\", \"that's it for today\" or similar). Load on sign-off and run the five parts: the task-log sweep, the memory sweep, the report, the note — always written — and the storyline fold, last.

How the storyline is written — the rolling summary of the notes that every session starts from, folded at session end. Load before every fold (summary_get → summary_set), and when the storyline reads as private shorthand.

A first tour of Agent Slate for a new user, run by the agent one step per turn — what each thing is, where to look in the browser, what changes. Load when the user asks for a tour or walkthrough, or is clearly new here.