Menu
System maintenance -- stale tasks, orphan folders, unsaved sessions, world health check
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Morning briefing -- read all walnut states, surface priorities, inbox count, stale walnuts, people nudges
Morning operating system -- sync inputs, read everything, surface priorities, show the day
Load an ALIVE walnut -- read kernel files, show current state, surface one observation, ask what to work on
Checkpoint -- route stash items, write log entry, update bundle, trigger projection, reset stash
Something new is emerging. A venture, an experiment, a person entering the orbit, a life area getting serious. It needs its own walnut — its own identity, history, and future. Scaffolds the full structure, maps existing context sources, and optionally migrates files across.
Generate a believable, lived-in ALIVE world from a free-text persona description (custom path) or a deterministic sandbox preset. Routes the create/list/activate/deactivate/delete/status surface and orchestrates the 5-stage subagent generation pipeline.
| name | alive-cleanup |
| description | System maintenance -- stale tasks, orphan folders, unsaved sessions, world health check |
| version | 1.0.0 |
| author | ALIVE Context System |
| license | MIT |
| toolsets | ["terminal","file"] |
| triggers | ["cleanup","system cleanup","tidy up","maintenance","health check"] |
| metadata | {"hermes":{"tags":["ALIVE","context","cleanup","maintenance"]}} |
Scan across all walnuts for entropy. Surface issues one at a time.
grep -l "saves: 0" "$ALIVE_WORLD_ROOT/.alive/_squirrels/"*.yaml 2>/dev/null
For each: check transcript file size before dismissing. saves: 0 + stash: [] does NOT mean empty -- the stash is only written at save.
[N] unsaved sessions found.
[id] -- [date] -- [walnut] -- transcript: [size]
Review transcripts for lost work?
1. Yes, check them all
2. Show me the list first
3. Clear and move on
Find bundles in draft status unchanged for 30+ days:
find "$ALIVE_WORLD_ROOT" -name "context.manifest.yaml" -mtime +30 2>/dev/null
For each stale bundle:
"[bundle-name]" has been in draft for [N] days.
1. Advance it (prototype/published)
2. Archive it (set status: done)
3. Kill it (delete -- drafts are disposable)
4. Leave it
Walnuts past 2x their rhythm with no recent sessions:
[walnut] -- waiting -- [N] days past weekly rhythm
1. Load it (check what's there)
2. Archive it (move to 01_Archive)
3. Update rhythm (maybe it's monthly now)
4. Leave it
Walnuts with stale or empty task files:
find "$ALIVE_WORLD_ROOT" -name "tasks.json" -exec python3 -c "
import json, sys
data = json.load(open(sys.argv[1]))
tasks = data.get('tasks', [])
active = [t for t in tasks if t.get('status') not in ('done', 'archived')]
if active:
print(f'{sys.argv[1]}: {len(active)} open tasks')
" {} \;
Look for v1/v2 patterns that should be migrated:
_core/ directories (v1)_kernel/_generated/ directories (v2)now.md files (v1)tasks.md files (v1)Found [N] walnuts with legacy structure.
1. Migrate now
2. Show me which ones
3. Skip
Surface issues one at a time. Don't overwhelm. Fix one, then ask if they want to continue.