Help us improve
Share bugs, ideas, or general feedback.
From cortex
Debug and fix memory system issues: validate memories, rate quality, manage protection, forget bad memories, and restore from checkpoints.
npx claudepluginhub cdeust/cortex --plugin cortexHow this skill is triggered — by the user, by Claude, or both
Slash command
/cortex:cortex-debug-memoryThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
debug, fix, bad memory, wrong, delete, forget, protect, rate, quality, restore, checkpoint, undo, stale, incorrect, mark useful, mark not useful, anchor, unprotect
Audits Pensyve memories for staleness (>30 days unaccessed or low retrievability), contradictions, low confidence (<0.5), and consolidation candidates, then offers confirmed cleanup actions. Use periodically.
Audits and selectively forgets stored Claude Code memories. Use when memory is large/uncurated, project state has shifted, or retrieval quality degraded.
Manages the total-recall memory system: captures, stores, retrieves, and forgets memories across tiers. Activates on /total-recall:commands or when users ask about memory, knowledge bases, or eval.
Share bugs, ideas, or general feedback.
debug, fix, bad memory, wrong, delete, forget, protect, rate, quality, restore, checkpoint, undo, stale, incorrect, mark useful, mark not useful, anchor, unprotect
Tools for maintaining memory quality — rate memories as useful or not, protect critical ones from decay, forget incorrect ones, validate against the filesystem, and restore from checkpoints when things go wrong.
Use this skill when: Recall returns wrong results, memories are stale, you need to undo changes, or you want to improve retrieval quality through feedback.
Soft delete (sets heat to 0, memory still exists but won't surface):
cortex:forget({
"memory_id": <id>,
"hard": false
})
Hard delete (permanent removal):
cortex:forget({
"memory_id": <id>,
"hard": true
})
Protected memories require "force": true to delete.
Provide feedback to train the metamemory confidence model:
cortex:rate_memory({
"memory_id": <id>,
"useful": true
})
Or mark as not useful:
cortex:rate_memory({
"memory_id": <id>,
"useful": false,
"reason": "outdated — we no longer use this approach"
})
Ratings adjust the memory's confidence score, which affects future retrieval ranking. Over time, this trains the system to surface better results.
Anchor a memory (heat=1.0 permanently, injected at session start):
cortex:anchor({
"memory_id": <id>,
"reason": "Core architecture decision — never decay"
})
Check if memories reference things that still exist:
cortex:validate_memory({
"directory": "<project root>"
})
Returns a list of stale memories (referencing deleted files, moved modules, etc.) that should be forgotten or updated.
Save a checkpoint before risky operations:
cortex:checkpoint({
"action": "save",
"label": "before-cleanup"
})
Restore if something went wrong:
cortex:checkpoint({
"action": "restore",
"label": "before-cleanup"
})
List available checkpoints:
cortex:checkpoint({
"action": "list"
})
Checkpoints are also created automatically before context compaction (via the compaction hook).
Recall returns irrelevant results:
useful: falseuseful: truevalidate_memory to find stale contentToo many memories on the same topic:
cortex:consolidate — CLS will merge similar episodic memories into semantic onesforget duplicatesMemory seems wrong/outdated:
Lost important context after compaction:
cortex:checkpoint({ "action": "list" }) for auto-checkpoints