From notably
Run the notably cleanup pass: hard-delete done todos older than clean.done_todo_age_days, interactively triage inactive reminders, and interactively triage stale non-permanent notes. Slash-only: runs when the user types /notably:clean.
How this skill is triggered — by the user, by Claude, or both
Slash command
/notably:cleanThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Three passes, in order:
Three passes, in order:
status: done whose created is older than clean.done_todo_age_days (default 7) is hard-deleted with no per-item prompt.status: inactive reminder, interactively triage via AskUserQuestion. Choices: keep, archive (move to ~/.notably/archive/reminders/ and tear down any remaining os or schedule entries), delete (same teardown).last_accessed is older than clean.note_stale_days (default 90), interactively triage. Choices: keep, archive, delete, mark permanent. If last_accessed is missing on a hand-edited note, created is used instead.node "${CLAUDE_PLUGIN_ROOT}/lib/node/cli.mjs" clean-scan
This emits a JSON object with three keys: deleted_done_todos (array of handles deleted automatically), inactive_reminders (array of records to triage), stale_notes (array of records to triage).
inactive_reminders, ask the user via AskUserQuestion with choices: keep / archive / delete. Then apply:node "${CLAUDE_PLUGIN_ROOT}/lib/node/cli.mjs" clean-apply <handle> <keep|archive|delete>
For each record in stale_notes, ask the user with choices: keep / archive / delete / mark_permanent. Apply via the same clean-apply subcommand.
Report a final summary: N todos deleted, N reminders archived/deleted/kept, N notes archived/deleted/marked permanent.
The split between clean-scan and clean-apply lets you drive AskUserQuestion interactively from this skill while keeping the CLI itself non-interactive.
npx claudepluginhub brennacodes/utilitry --plugin notablyGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.