← ClaudeAtlas

prunelisted

Bulk-delete finished Ogre runtime data (rescue logs/tmp by default, optionally whole completed/stopped feature issues too) instead of targeting one issue at a time with /ogre:stop --delete.
metallurgical/ogre-runner · ★ 1 · AI & Automation · score 64
Install: claude install-skill metallurgical/ogre-runner
# /ogre:prune Use this skill when the user wants to clean up accumulated Ogre runtime data in bulk - most commonly the ad-hoc `.ai/.ogre/{logs,tmp}/issue-rescue-<slug>/` directories every `/ogre:rescue` call creates, which have no other cleanup path (`/ogre:stop --delete` targets one issue at a time, and nothing runs automatically). Alias: `/ogre:purge` (identical, same flags). ## Inputs - `--all` - also sweeps completed/stopped feature/execute issues, not just rescues. Without it, only `rescue-*` ledger tasks are ever considered. - `--older-than N` - age safety margin in days (default `1`). Only tasks/ issues finished more than N days ago are eligible. `--older-than 0` overrides for immediate cleanup. - `--yes` - actually deletes. Without it, `ogre prune` only previews what's eligible and deletes nothing (dry-run is the default, not opt-in). ## Behavior Run: - `${CLAUDE_PLUGIN_ROOT}/scripts/ogre prune [--all] [--older-than N] [--yes]` Two-step by default - **always run it once without `--yes` first and show the user the preview** (issue, finished date, reclaimable size) before adding `--yes`. This is a bulk-delete operation; do not skip straight to `--yes` just because the user said "clean it up" or "prune the old stuff" - show them what's eligible, get their go-ahead on the specific list (not just the idea in the abstract), then re-run with `--yes`. Exception: if the user's own message already specifies age and scope precisely enough that the eligible set is